{"idx": 0, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ro7tuDVP6Ks.mp4", "duration": 82.0825, "prompt_text": "Please find the visual event described by the sentence 'When is a group of people wearing white robes walking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a group of people wearing white robes walking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 15.0], "base_se_norm": [-0.9512685537338257, -0.6345140337944031], "gt_se_raw": [3.0, 9.0], "gt_se_norm": [-0.9269028306007385, -0.7807084321975708], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000000.pt"}
{"idx": 2, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ro7tuDVP6Ks.mp4", "duration": 82.0825, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman in a black jacket hugs a woman in a white coat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman in a black jacket hugs a woman in a white coat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.025370776653289795, 0.04772639274597168], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.0010050535202026367, 0.04772639274597168], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000002.pt"}
{"idx": 4, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ro7tuDVP6Ks.mp4", "duration": 82.0825, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the students are standing on stage holding candles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the students are standing on stage holding candles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 72.0], "base_se_norm": [0.6081382036209106, 0.7543326616287231], "gt_se_raw": [67.0, 70.0], "gt_se_norm": [0.6325039863586426, 0.7056010961532593], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000004.pt"}
{"idx": 6, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdwFR_74t_0.mp4", "duration": 12.654, "prompt_text": "Please find the visual event described by the sentence 'At what moment do the colorful blocks spill out onto the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment do the colorful blocks spill out onto the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.5805279016494751, 0.8966333866119385], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.20973604917526245, -0.05168330669403076], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000006.pt"}
{"idx": 8, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdwFR_74t_0.mp4", "duration": 12.654, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a black top lean down to help with the mess?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a black top lean down to help with the mess?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [0.42247509956359863, 0.8966333866119385], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [0.42247509956359863, 0.7385807037353516], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000008.pt"}
{"idx": 10, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BeiRM6-PdOM.mp4", "duration": 335.923078, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the gray shirt being interviewed inside the hangar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the gray shirt being interviewed inside the hangar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 48.0], "base_se_norm": [-0.8928326368331909, -0.7142202854156494], "gt_se_raw": [19.0, 48.0], "gt_se_norm": [-0.8868788480758667, -0.7142202854156494], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000010.pt"}
{"idx": 12, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BeiRM6-PdOM.mp4", "duration": 335.923078, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man in sunglasses gives a thumbs-up while freefalling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 191.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man in sunglasses gives a thumbs-up while freefalling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 191.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [191.0, 193.0], "base_se_norm": [0.13716518878936768, 0.14907264709472656], "gt_se_raw": [194.0, 197.0], "gt_se_norm": [0.15502631664276123, 0.17288756370544434], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000012.pt"}
{"idx": 14, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BeiRM6-PdOM.mp4", "duration": 335.923078, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the skydivers land on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 271.0 - 302.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the skydivers land on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 271.0 - 302.0 seconds.<|im_end|>\n", "base_se_raw": [271.0, 302.0], "base_se_norm": [0.6134647130966187, 0.7980307340621948], "gt_se_raw": [300.0, 303.0], "gt_se_norm": [0.7861232757568359, 0.803984522819519], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000014.pt"}
{"idx": 16, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jIi4Fe6jRHo.mp4", "duration": 57.771244, "prompt_text": "Please find the visual event described by the sentence 'A soldier in uniform shoots a machine gun towards the waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A soldier in uniform shoots a machine gun towards the waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [-0.6191877126693726, -0.3768526315689087], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.6191877126693726, -0.3768526315689087], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000016.pt"}
{"idx": 18, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jIi4Fe6jRHo.mp4", "duration": 57.771244, "prompt_text": "Please find the visual event described by the sentence 'A soldier is walking in the river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A soldier is walking in the river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [0.35015273094177246, 0.45401060581207275], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [0.35015273094177246, 0.45401060581207275], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000018.pt"}
{"idx": 20, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pK-7v-CimFo.mp4", "duration": 60.07, "prompt_text": "Please find the visual event described by the sentence 'Show the process of shredding the potatoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of shredding the potatoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.8668220043182373, -0.7003495693206787], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.8335275650024414, -0.7336440682411194], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000020.pt"}
{"idx": 22, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pK-7v-CimFo.mp4", "duration": 60.07, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the hash brown mixture is placed into the waffle maker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the hash brown mixture is placed into the waffle maker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 41.0], "base_se_norm": [0.1986016035079956, 0.3650740385055542], "gt_se_raw": [36.0, 41.0], "gt_se_norm": [0.1986016035079956, 0.3650740385055542], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000022.pt"}
{"idx": 24, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pK-7v-CimFo.mp4", "duration": 60.07, "prompt_text": "Please find the visual event described by the sentence 'At what time do we see the runny yolk from the egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do we see the runny yolk from the egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 56.0], "base_se_norm": [0.7646080255508423, 0.8644914627075195], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.7979024648666382, 0.8644914627075195], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000024.pt"}
{"idx": 26, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x9FzWnWW95U.mp4", "duration": 10.541797, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man in a red shirt patterned with skulls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man in a red shirt patterned with skulls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.8102790117263794, 0.5177677869796753], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.8102790117263794, 0.5177677869796753], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000026.pt"}
{"idx": 28, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x9FzWnWW95U.mp4", "duration": 10.541797, "prompt_text": "Please find the visual event described by the sentence 'When does the man in glasses lean forward in his chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in glasses lean forward in his chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.6205580234527588, 0.13832592964172363], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.4308370351791382, -0.05139511823654175], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000028.pt"}
{"idx": 30, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yH2wT5aTmb8.mp4", "duration": 31.833333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the baby actively interacting with the blue book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the baby actively interacting with the blue book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.43455493450164795], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8115183115005493], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000030.pt"}
{"idx": 32, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yH2wT5aTmb8.mp4", "duration": 31.833333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman cover an object with the star-patterned blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman cover an object with the star-patterned blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.30890047550201416, -0.12041878700256348], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.30890047550201416, -0.12041878700256348], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000032.pt"}
{"idx": 34, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yH2wT5aTmb8.mp4", "duration": 31.833333, "prompt_text": "Please find the visual event described by the sentence 'When does the baby turn its head to look to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby turn its head to look to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.12041878700256348, 0.0052356719970703125], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [0.759162425994873, 0.9476441144943237], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000034.pt"}
{"idx": 36, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qniwI2hNhDs.mp4", "duration": 165.163333, "prompt_text": "Please find the visual event described by the sentence 'At what point do the sheep gather to form the shape of a single, large sheep?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the sheep gather to form the shape of a single, large sheep?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 46.0], "base_se_norm": [-0.5277402400970459, -0.44297564029693604], "gt_se_raw": [39.0, 46.0], "gt_se_norm": [-0.5277402400970459, -0.44297564029693604], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000036.pt"}
{"idx": 38, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qniwI2hNhDs.mp4", "duration": 165.163333, "prompt_text": "Please find the visual event described by the sentence 'When do the sheep with lights create a firework-like explosion in the dark?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the sheep with lights create a firework-like explosion in the dark?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 133.0], "base_se_norm": [0.549980878829956, 0.6105270385742188], "gt_se_raw": [129.0, 132.0], "gt_se_norm": [0.5620900392532349, 0.5984177589416504], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000038.pt"}
{"idx": 40, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uwFhl9lG-z0.mp4", "duration": 443.826211, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker introduce her second point about vague directions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker introduce her second point about vague directions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 122.0], "base_se_norm": [-0.495297908782959, -0.4502352476119995], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [-0.5583857297897339, -0.5448669195175171], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000040.pt"}
{"idx": 42, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uwFhl9lG-z0.mp4", "duration": 443.826211, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where she mentions the complaint about having to wear glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where she mentions the complaint about having to wear glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 120.0], "base_se_norm": [-0.49980419874191284, -0.45924776792526245], "gt_se_raw": [216.0, 224.0], "gt_se_norm": [-0.026646018028259277, 0.009404182434082031], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000042.pt"}
{"idx": 44, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uwFhl9lG-z0.mp4", "duration": 443.826211, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the speaker points her finger to demonstrate a vague direction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the speaker points her finger to demonstrate a vague direction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 113.0], "base_se_norm": [-0.49980419874191284, -0.4907916784286499], "gt_se_raw": [111.0, 112.0], "gt_se_norm": [-0.49980419874191284, -0.495297908782959], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000044.pt"}
{"idx": 46, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-CR65sS1Frw.mp4", "duration": 5.410233, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the explosion expands into a large fireball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the explosion expands into a large fireball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.2606602907180786], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.6303301453590393, -0.2606602907180786], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000046.pt"}
{"idx": 48, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-CR65sS1Frw.mp4", "duration": 5.410233, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range where flames are visible on the car's roof after the main explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range where flames are visible on the car's roof after the main explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.6303301453590393, 0.1090095043182373], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.2606602907180786, 0.4786794185638428], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000048.pt"}
{"idx": 50, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KxwbAqLb5Qk.mp4", "duration": 105.743, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker talk about why they are called pancake tortoises?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker talk about why they are called pancake tortoises?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 49.0], "base_se_norm": [-0.6406381130218506, -0.07322466373443604], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.7730346322059631, -0.7162932753562927], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000050.pt"}
{"idx": 52, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KxwbAqLb5Qk.mp4", "duration": 105.743, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the speaker explains that pancake tortoises are very fast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the speaker explains that pancake tortoises are very fast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [-0.5271554589271545, -0.28127628564834595], "gt_se_raw": [27.0, 35.0], "gt_se_norm": [-0.48932790756225586, -0.33801764249801636], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000052.pt"}
{"idx": 54, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KxwbAqLb5Qk.mp4", "duration": 105.743, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the major threats to pancake tortoises are listed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the major threats to pancake tortoises are listed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 89.0], "base_se_norm": [0.22939586639404297, 0.6833266019821167], "gt_se_raw": [69.0, 91.0], "gt_se_norm": [0.30505096912384033, 0.7211542129516602], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000054.pt"}
{"idx": 56, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iimJIWQqAGA.mp4", "duration": 58.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl press a button to select a balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl press a button to select a balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.34527915716171265, -0.2419021725654602], "gt_se_raw": [20.0, 21.0], "gt_se_norm": [-0.31082016229629517, -0.2763611674308777], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000056.pt"}
{"idx": 58, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iimJIWQqAGA.mp4", "duration": 58.04, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the girl takes the balloon from the dispenser.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the girl takes the balloon from the dispenser.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 35.0], "base_se_norm": [0.033769845962524414, 0.20606482028961182], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [0.0682288408279419, 0.20606482028961182], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000058.pt"}
{"idx": 60, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3xPDE8H0Jw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When are two construction workers seen on scaffolding working on a house roof?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two construction workers seen on scaffolding working on a house roof?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.4753605127334595], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000060.pt"}
{"idx": 62, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3xPDE8H0Jw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a walkthrough of the wooden framework inside the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a walkthrough of the wooden framework inside the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.016300976276397705, 0.44275856018066406], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000062.pt"}
{"idx": 64, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3xPDE8H0Jw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man in a plaid shirt is working on the roof of a house under construction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man in a plaid shirt is working on the roof of a house under construction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.606520414352417], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.4753605127334595], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000064.pt"}
{"idx": 66, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4Pr7PZ3Bgc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point are two men practicing in a virtual shooting range?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two men practicing in a virtual shooting range?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.47478586435317993, -0.21217882633209229], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000066.pt"}
{"idx": 68, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4Pr7PZ3Bgc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a young woman is carefully dusting a blue handgun for fingerprints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a young woman is carefully dusting a blue handgun for fingerprints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4443387985229492, 0.8382494449615479], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [0.4443387985229492, 0.7069458961486816], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000068.pt"}
{"idx": 70, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RsZE4a4ol_U.mp4", "duration": 129.06226666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the exterior of the building with a statue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the exterior of the building with a statue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 123.0], "base_se_norm": [0.8440710306167603, 0.9060566425323486], "gt_se_raw": [118.0, 120.0], "gt_se_norm": [0.8285746574401855, 0.859567403793335], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000070.pt"}
{"idx": 72, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RsZE4a4ol_U.mp4", "duration": 129.06226666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are we shown a bathroom with two sinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are we shown a bathroom with two sinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [-0.2096838355064392, -0.1631946563720703], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.2096838355064392, -0.1631946563720703], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000072.pt"}
{"idx": 74, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RsZE4a4ol_U.mp4", "duration": 129.06226666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the tour of the empty double room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the tour of the empty double room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 69.0], "base_se_norm": [-0.14769822359085083, 0.06925129890441895], "gt_se_raw": [55.0, 69.0], "gt_se_norm": [-0.14769822359085083, 0.06925129890441895], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000074.pt"}
{"idx": 76, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VyHV0BRtdxo.mp4", "duration": 136.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does Hagrid tell Harry he's a wizard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Hagrid tell Harry he's a wizard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.8678494691848755, -0.794432520866394], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [-0.7797491550445557, -0.7650657892227173], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000076.pt"}
{"idx": 78, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VyHV0BRtdxo.mp4", "duration": 136.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does Hermione cast the Petrificus Totalus spell?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Hermione cast the Petrificus Totalus spell?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.3979809880256653, -0.35393083095550537], "gt_se_raw": [80.0, 82.0], "gt_se_norm": [0.17467117309570312, 0.20403802394866943], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000078.pt"}
{"idx": 80, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9-k5J4RxQdE.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'When does Daffy Duck first appear in the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Daffy Duck first appear in the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9076923131942749, -0.8153846263885498], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.9076923131942749, -0.8769230842590332], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000080.pt"}
{"idx": 82, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9-k5J4RxQdE.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where Daffy is wearing a red wig and talking to Bugs Bunny.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where Daffy is wearing a red wig and talking to Bugs Bunny.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 58.0], "base_se_norm": [-0.01538461446762085, 0.7846153974533081], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [-0.04615384340286255, 0.015384674072265625], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000082.pt"}
{"idx": 84, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9-k5J4RxQdE.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when Daffy grabs his own beak and twists it back into place.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when Daffy grabs his own beak and twists it back into place.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.23076921701431274, -0.04615384340286255], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [0.6000000238418579, 0.6615384817123413], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000084.pt"}
{"idx": 86, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/skKUzMST92g.mp4", "duration": 144.0, "prompt_text": "Please find the visual event described by the sentence 'Show the demonstration of the knit two, purl two pattern.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the demonstration of the knit two, purl two pattern.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 69.0], "base_se_norm": [-0.8194444179534912, -0.04166668653488159], "gt_se_raw": [19.0, 31.0], "gt_se_norm": [-0.7361111044883728, -0.5694444179534912], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000086.pt"}
{"idx": 88, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/skKUzMST92g.mp4", "duration": 144.0, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a locking stitch marker is being clipped onto the knitted work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a locking stitch marker is being clipped onto the knitted work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 89.0], "base_se_norm": [-0.04166668653488159, 0.23611116409301758], "gt_se_raw": [68.0, 84.0], "gt_se_norm": [-0.05555558204650879, 0.16666662693023682], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000088.pt"}
{"idx": 90, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xn_XpOYgUi8.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the tea leaves added to the boiling water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the tea leaves added to the boiling water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.19853240251541138, -0.10138481855392456], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.17424547672271729, -0.10138481855392456], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000090.pt"}
{"idx": 92, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xn_XpOYgUi8.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is milk added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is milk added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 51.0], "base_se_norm": [0.14148414134979248, 0.2386317253112793], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [0.1657710075378418, 0.2386317253112793], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000092.pt"}
{"idx": 94, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xn_XpOYgUi8.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip displaying the ingredients needed for two cups of tea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip displaying the ingredients needed for two cups of tea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 26.0], "base_se_norm": [-0.4899751543998718, -0.3685407042503357], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.4899751543998718, -0.3685407042503357], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000094.pt"}
{"idx": 96, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vg3Q2W-oGgE.mp4", "duration": 147.01353333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the girl walks to the kitchen area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the girl walks to the kitchen area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.22456121444702148, -0.14293605089187622], "gt_se_raw": [54.0, 65.0], "gt_se_norm": [-0.26537376642227173, -0.11572766304016113], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000096.pt"}
{"idx": 98, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vg3Q2W-oGgE.mp4", "duration": 147.01353333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where she walks back to her desk with the yarn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where she walks back to her desk with the yarn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 114.0], "base_se_norm": [-0.26537376642227173, 0.550877571105957], "gt_se_raw": [106.0, 112.0], "gt_se_norm": [0.44204413890838623, 0.5236692428588867], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000098.pt"}
{"idx": 100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-E3cRZCne4.mp4", "duration": 75.04, "prompt_text": "Please find the visual event described by the sentence 'When does the video mention '670 million jobs'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video mention '670 million jobs'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.7601279020309448, -0.6535181403160095], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.7334754467010498, -0.6535181403160095], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000100.pt"}
{"idx": 102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-E3cRZCne4.mp4", "duration": 75.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip that discusses reason number 4.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip that discusses reason number 4.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 47.0], "base_se_norm": [-0.06716418266296387, 0.25266528129577637], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.06716418266296387, 0.012793183326721191], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000102.pt"}
{"idx": 104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-E3cRZCne4.mp4", "duration": 75.04, "prompt_text": "Please find the visual event described by the sentence 'When can we see the website address at the end of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the website address at the end of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 69.0], "base_se_norm": [0.6524519920349121, 0.8390191793441772], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.6524519920349121, 0.7857142686843872], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000104.pt"}
{"idx": 106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_ikKgIuqVX4.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show me the part of the video where the speaker provides a general summary about the employment disparity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part of the video where the speaker provides a general summary about the employment disparity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.22407686710357666, 0.35786545276641846], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000106.pt"}
{"idx": 108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_ikKgIuqVX4.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the speaker begins the narration with 'As you can see'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the speaker begins the narration with 'As you can see'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.8060191869735718], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.9030095934867859], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000108.pt"}
{"idx": 110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QnrNPVaeDMk.mp4", "duration": 100.13336666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man help the boy with his hearing aids?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man help the boy with his hearing aids?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 65.0], "base_se_norm": [-0.9800266623497009, 0.29826855659484863], "gt_se_raw": [13.0, 20.0], "gt_se_norm": [-0.7403463125228882, -0.6005327701568604], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000110.pt"}
{"idx": 112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QnrNPVaeDMk.mp4", "duration": 100.13336666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the man and the boy are shaking the bottle of mousse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the man and the boy are shaking the bottle of mousse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 49.0], "base_se_norm": [-0.1411454677581787, -0.02130526304244995], "gt_se_raw": [44.0, 49.0], "gt_se_norm": [-0.12117207050323486, -0.02130526304244995], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000112.pt"}
{"idx": 114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QnrNPVaeDMk.mp4", "duration": 100.13336666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the timeframe where the boy reaches into the leftmost orange compartment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the timeframe where the boy reaches into the leftmost orange compartment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.8601864576339722, -0.7802929878234863], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.8601864576339722, -0.7603196501731873], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000114.pt"}
{"idx": 116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dev6T3ZCrY8.mp4", "duration": 34.034, "prompt_text": "Please find the visual event described by the sentence 'When is the dog standing on the bow of a ship surrounded by ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog standing on the bow of a ship surrounded by ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.6474114060401917, -0.5298818945884705], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.6474114060401917, -0.5298818945884705], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000116.pt"}
{"idx": 118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dev6T3ZCrY8.mp4", "duration": 34.034, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot of a man and a dog looking out over a vast green valley at sunrise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot of a man and a dog looking out over a vast green valley at sunrise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 25.0], "base_se_norm": [0.23406004905700684, 0.4691190719604492], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [0.23406004905700684, 0.4691190719604492], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000118.pt"}
{"idx": 120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EMJ0kz8Iu9o.mp4", "duration": 54.320933333333336, "prompt_text": "Please find the visual event described by the sentence 'When are the large yellow cranes being assembled on the construction site?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the large yellow cranes being assembled on the construction site?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.5581814050674438, -0.26363563537597656], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.5581814050674438, -0.4477267265319824], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000120.pt"}
{"idx": 122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EMJ0kz8Iu9o.mp4", "duration": 54.320933333333336, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the initial ground excavation and foundation preparation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the initial ground excavation and foundation preparation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 12.0], "base_se_norm": [-1.0, -0.5581814050674438], "gt_se_raw": [5.0, 12.0], "gt_se_norm": [-0.8159089088439941, -0.5581814050674438], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000122.pt"}
{"idx": 124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EMJ0kz8Iu9o.mp4", "duration": 54.320933333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see a concrete pump truck with a long red boom pouring concrete?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see a concrete pump truck with a long red boom pouring concrete?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 34.0], "base_se_norm": [-0.0795445442199707, 0.251819372177124], "gt_se_raw": [25.0, 32.0], "gt_se_norm": [-0.0795445442199707, 0.1781829595565796], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000124.pt"}
{"idx": 126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3_TG_mMywxE.mp4", "duration": 89.54779166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the clip that shows a person measuring a piece of wood with a ruler.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip that shows a person measuring a piece of wood with a ruler.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [-0.3523011803627014, -0.3076322674751282], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.3969700336456299, -0.3523011803627014], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000126.pt"}
{"idx": 128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3_TG_mMywxE.mp4", "duration": 89.54779166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me when the tablet displays a 45-degree miter angle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the tablet displays a 45-degree miter angle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [0.20605993270874023, 0.27306318283081055], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [0.22839438915252686, 0.27306318283081055], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000128.pt"}
{"idx": 130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RzR_O2DoSVs.mp4", "duration": 29.229200000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point does the adult guide the child's hand to repeatedly pat the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the adult guide the child's hand to repeatedly pat the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 15.0], "base_se_norm": [-0.6578763723373413, 0.026370882987976074], "gt_se_raw": [6.0, 15.0], "gt_se_norm": [-0.5894516706466675, 0.026370882987976074], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000130.pt"}
{"idx": 132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RzR_O2DoSVs.mp4", "duration": 29.229200000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the adult demonstrates how to play the drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the adult demonstrates how to play the drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.026370882987976074], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.7263010740280151], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000132.pt"}
{"idx": 134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RzR_O2DoSVs.mp4", "duration": 29.229200000000002, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the baby is watching the adult and has its finger in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the baby is watching the adult and has its finger in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.026370882987976074], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.5894516706466675], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000134.pt"}
{"idx": 136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/18oCRSIVeec.mp4", "duration": 101.20240480961924, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the student walks down a hallway with other people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the student walks down a hallway with other people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [-0.34784162044525146, -0.24902969598770142], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.3280792236328125, -0.24902969598770142], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000136.pt"}
{"idx": 138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/18oCRSIVeec.mp4", "duration": 101.20240480961924, "prompt_text": "Please find the visual event described by the sentence 'Show the exterior shot of the brick college building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the exterior shot of the brick college building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 64.0], "base_se_norm": [0.2055048942565918, 0.2647920846939087], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [0.2055048942565918, 0.24502968788146973], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000138.pt"}
{"idx": 140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_iU_gnn28U.mp4", "duration": 190.12326666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a teacher instructing a young apprentice on the code of the samurai?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a teacher instructing a young apprentice on the code of the samurai?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 32.0], "base_se_norm": [-0.8948050737380981, -0.6633762717247009], "gt_se_raw": [12.0, 32.0], "gt_se_norm": [-0.8737661242485046, -0.6633762717247009], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000140.pt"}
{"idx": 142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_iU_gnn28U.mp4", "duration": 190.12326666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the protagonist perform a stealth takedown by breaking through a paper wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the protagonist perform a stealth takedown by breaking through a paper wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 154.0], "base_se_norm": [0.5884431600570679, 0.6200016736984253], "gt_se_raw": [153.0, 155.0], "gt_se_norm": [0.6094821691513062, 0.6305211782455444], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000142.pt"}
{"idx": 144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_iU_gnn28U.mp4", "duration": 190.12326666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the warrior riding a horse through a forest with white-leafed trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the warrior riding a horse through a forest with white-leafed trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 151.0], "base_se_norm": [0.5674042701721191, 0.5884431600570679], "gt_se_raw": [149.0, 152.0], "gt_se_norm": [0.5674042701721191, 0.598962664604187], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000144.pt"}
{"idx": 146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MizXqgEuhe4.mp4", "duration": 77.04363333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the girl in the blue shirt is speaking into the microphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the girl in the blue shirt is speaking into the microphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 73.0], "base_se_norm": [-0.896162748336792, 0.8950300216674805], "gt_se_raw": [4.0, 73.0], "gt_se_norm": [-0.896162748336792, 0.8950300216674805], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000146.pt"}
{"idx": 148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MizXqgEuhe4.mp4", "duration": 77.04363333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman sitting to the right of the speaker is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman sitting to the right of the speaker is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 73.0], "base_se_norm": [-0.792325496673584, 0.8950300216674805], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.792325496673584, -0.688488245010376], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000148.pt"}
{"idx": 150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKIIf6W16Uk.mp4", "duration": 276.6433099766433, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman in a white jacket is introducing the museum to a man in a black hoodie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman in a white jacket is introducing the museum to a man in a black hoodie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 65.0], "base_se_norm": [-0.6963598728179932, -0.5300807952880859], "gt_se_raw": [42.0, 58.0], "gt_se_norm": [-0.6963598728179932, -0.5806875228881836], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000150.pt"}
{"idx": 152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKIIf6W16Uk.mp4", "duration": 276.6433099766433, "prompt_text": "Please find the visual event described by the sentence 'When does the man have a flashback to his childhood where he is ignored by an ice cream truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man have a flashback to his childhood where he is ignored by an ice cream truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 158.0], "base_se_norm": [-0.00232541561126709, 0.14226508140563965], "gt_se_raw": [138.0, 159.0], "gt_se_norm": [-0.00232541561126709, 0.14949464797973633], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000152.pt"}
{"idx": 154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKIIf6W16Uk.mp4", "duration": 276.6433099766433, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man, holding a soft-serve ice cream, is talking to the woman in the gift shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 248.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man, holding a soft-serve ice cream, is talking to the woman in the gift shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 248.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 248.0], "base_se_norm": [0.6121842861175537, 0.7929224967956543], "gt_se_raw": [225.0, 247.0], "gt_se_norm": [0.6266433000564575, 0.7856929302215576], "start_token_span": [81, 86], "end_token_span": [88, 93], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000154.pt"}
{"idx": 156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VfDWQG47pAQ.mp4", "duration": 177.24373333333332, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man in the blue shirt talk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man in the blue shirt talk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8871610164642334, -0.8533093333244324], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.8871610164642334, -0.8533093333244324], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000156.pt"}
{"idx": 158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VfDWQG47pAQ.mp4", "duration": 177.24373333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the girl walk through the snowy forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl walk through the snowy forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 96.0], "base_se_norm": [0.004266858100891113, 0.08325409889221191], "gt_se_raw": [89.0, 95.0], "gt_se_norm": [0.004266858100891113, 0.07197022438049316], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000158.pt"}
{"idx": 160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_WrjdMns5Vc.mp4", "duration": 19.227541666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see the clock on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the clock on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 12.0], "base_se_norm": [-0.895982563495636, 0.24820959568023682], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.895982563495636, -0.6879476308822632], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000160.pt"}
{"idx": 162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_WrjdMns5Vc.mp4", "duration": 19.227541666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'NEED HELP MANAGING STRESS?' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'NEED HELP MANAGING STRESS?' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [0.35222697257995605, 0.5602619647979736], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [0.45624446868896484, 0.6642793416976929], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000162.pt"}
{"idx": 164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_WrjdMns5Vc.mp4", "duration": 19.227541666666664, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the student is writing with a green pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the student is writing with a green pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.6879476308822632, 0.35222697257995605], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.6879476308822632, -0.4799126982688904], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000164.pt"}
{"idx": 166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-Z8xxygd2Y.mp4", "duration": 105.76, "prompt_text": "Please find the visual event described by the sentence 'A woman is sleeping on a golden bed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is sleeping on a golden bed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 37.0], "base_se_norm": [-0.6785174012184143, -0.30030256509780884], "gt_se_raw": [17.0, 27.0], "gt_se_norm": [-0.6785174012184143, -0.4894099831581116], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000166.pt"}
{"idx": 168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-Z8xxygd2Y.mp4", "duration": 105.76, "prompt_text": "Please find the visual event described by the sentence 'The woman wakes up and smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The woman wakes up and smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 74.0], "base_se_norm": [0.1913766860961914, 0.3993948698043823], "gt_se_raw": [68.0, 74.0], "gt_se_norm": [0.2859303951263428, 0.3993948698043823], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000168.pt"}
{"idx": 170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H7pK1f-mtoI.mp4", "duration": 69.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman tickling the boy under his chin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman tickling the boy under his chin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.02567160129547119, 0.14626872539520264], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.7707462310791016, -0.713432788848877], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000170.pt"}
{"idx": 172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H7pK1f-mtoI.mp4", "duration": 69.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman is holding all five of the boy's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman is holding all five of the boy's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.5128358006477356, -0.34089547395706177], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.4841790795326233, -0.3695521950721741], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000172.pt"}
{"idx": 174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H7pK1f-mtoI.mp4", "duration": 69.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman touches the boy's thumb.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman touches the boy's thumb.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 37.0], "base_se_norm": [-0.02567160129547119, 0.06029856204986572], "gt_se_raw": [66.0, 68.0], "gt_se_norm": [0.891343355178833, 0.9486567974090576], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000174.pt"}
{"idx": 176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g2wsNw07wRY.mp4", "duration": 132.46566666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a white dog runs along the edge of the pool with a red toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a white dog runs along the edge of the pool with a red toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.7131332159042358, -0.6527402400970459], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [-0.0639083981513977, -0.033711910247802734], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000176.pt"}
{"idx": 178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g2wsNw07wRY.mp4", "duration": 132.46566666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a dog is playing with a bucket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a dog is playing with a bucket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 26.0], "base_se_norm": [-0.6829367280006409, -0.607445478439331], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.6829367280006409, -0.6527402400970459], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000178.pt"}
{"idx": 180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tT7jelfqCCw.mp4", "duration": 80.34693333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a classroom with two large projection screens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a classroom with two large projection screens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.6266192197799683, -0.5270509719848633], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.6017271280288696, -0.5270509719848633], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000180.pt"}
{"idx": 182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tT7jelfqCCw.mp4", "duration": 80.34693333333334, "prompt_text": "Please find the visual event described by the sentence 'When are three students seen laughing together while looking at a book with a stained glass window in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three students seen laughing together while looking at a book with a stained glass window in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 45.0], "base_se_norm": [-0.02920997142791748, 0.12014234066009521], "gt_se_raw": [40.0, 45.0], "gt_se_norm": [-0.004317939281463623, 0.12014234066009521], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000182.pt"}
{"idx": 184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tT7jelfqCCw.mp4", "duration": 80.34693333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a checkered shirt shown looking at a book with a female student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a checkered shirt shown looking at a book with a female student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 49.0], "base_se_norm": [0.12014234066009521, 0.2197105884552002], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [0.12014234066009521, 0.2197105884552002], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000184.pt"}
{"idx": 186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tODlPxhbGF8.mp4", "duration": 144.35, "prompt_text": "Please find the visual event described by the sentence 'At what point do the fans in the red shirts start jumping up and down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the fans in the red shirts start jumping up and down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 58.0], "base_se_norm": [-0.3765154480934143, -0.19639766216278076], "gt_se_raw": [43.0, 106.0], "gt_se_norm": [-0.4042258858680725, 0.46865248680114746], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000186.pt"}
{"idx": 188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tODlPxhbGF8.mp4", "duration": 144.35, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when a player is tackled to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when a player is tackled to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 138.0], "base_se_norm": [0.8427432775497437, 0.9120193719863892], "gt_se_raw": [135.0, 137.0], "gt_se_norm": [0.8704537153244019, 0.8981641530990601], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000188.pt"}
{"idx": 190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1TE0HIGNStc.mp4", "duration": 151.61813333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the blue shirt applying lip gloss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the blue shirt applying lip gloss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.8417075872421265, -0.7757524251937866], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.8021345138549805, -0.7625613808631897], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000190.pt"}
{"idx": 192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1TE0HIGNStc.mp4", "duration": 151.61813333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man turn his head to look at the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man turn his head to look at the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 50.0], "base_se_norm": [-0.3932124376296997, -0.340448260307312], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [-0.18215584754943848, -0.14258277416229248], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000192.pt"}
{"idx": 194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1TE0HIGNStc.mp4", "duration": 151.61813333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the plaid shirt exit the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the plaid shirt exit the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 103.0], "base_se_norm": [0.29272139072418213, 0.358676552772522], "gt_se_raw": [97.0, 103.0], "gt_se_norm": [0.27953028678894043, 0.358676552772522], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000194.pt"}
{"idx": 196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bDird_PgnVs.mp4", "duration": 42.867000000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the teacher guides the boy's hand to touch the blue rectangular objects.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the teacher guides the boy's hand to touch the blue rectangular objects.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.7667202949523926, -0.5800966024398804], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.7667202949523926, -0.5800966024398804], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000196.pt"}
{"idx": 198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bDird_PgnVs.mp4", "duration": 42.867000000000004, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman guides the boy's hand to touch the symbol for breakfast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman guides the boy's hand to touch the symbol for breakfast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.8600322008132935, -0.5800966024398804], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [0.4929898977279663, 0.7262696027755737], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000198.pt"}
{"idx": 200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fZ9WiuJPnNA.mp4", "duration": 140.50703333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see four penguin puppets peeking out from behind a door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see four penguin puppets peeking out from behind a door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 58.0], "base_se_norm": [-0.37369686365127563, -0.17441856861114502], "gt_se_raw": [45.0, 59.0], "gt_se_norm": [-0.3594626784324646, -0.16018438339233398], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000200.pt"}
{"idx": 202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fZ9WiuJPnNA.mp4", "duration": 140.50703333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman singing with four chickens wearing hats and sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman singing with four chickens wearing hats and sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 88.0], "base_se_norm": [0.010625600814819336, 0.25260627269744873], "gt_se_raw": [71.0, 88.0], "gt_se_norm": [0.010625600814819336, 0.25260627269744873], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000202.pt"}
{"idx": 204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fZ9WiuJPnNA.mp4", "duration": 140.50703333333334, "prompt_text": "Please find the visual event described by the sentence 'When are all the different types of puppets (monsters, penguins, and chickens) singing together in the same shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the different types of puppets (monsters, penguins, and chickens) singing together in the same shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 116.0], "base_se_norm": [0.4376504421234131, 0.65116286277771], "gt_se_raw": [97.0, 110.0], "gt_se_norm": [0.3807138204574585, 0.5657578706741333], "start_token_span": [80, 85], "end_token_span": [87, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000204.pt"}
{"idx": 206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iYdb1mr7YhQ.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a student is studying in the library.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a student is studying in the library.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000206.pt"}
{"idx": 208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iYdb1mr7YhQ.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the text 'YOU'VE GOT THIS!' is displayed on a blue background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the text 'YOU'VE GOT THIS!' is displayed on a blue background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5066900253295898, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000208.pt"}
{"idx": 210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I_D2UbSnkFg.mp4", "duration": 81.8818, "prompt_text": "Please find the visual event described by the sentence 'When is the woman feeding the boy with a spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman feeding the boy with a spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 39.0], "base_se_norm": [-0.9755745530128479, -0.047407329082489014], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9022982120513916], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000210.pt"}
{"idx": 212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I_D2UbSnkFg.mp4", "duration": 81.8818, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the boy opens his mouth as if to speak or eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the boy opens his mouth as if to speak or eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.1695346236228943, -0.07183277606964111], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.1695346236228943, -0.12068367004394531], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000212.pt"}
{"idx": 214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I_D2UbSnkFg.mp4", "duration": 81.8818, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the boy is wearing a neck brace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the boy is wearing a neck brace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 81.0], "base_se_norm": [-1.0, 0.978461742401123], "gt_se_raw": [0.0, 81.0], "gt_se_norm": [-1.0, 0.978461742401123], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000214.pt"}
{"idx": 216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hgLQQe5uUCE.mp4", "duration": 89.75633333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person walk across the street towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person walk across the street towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 32.0], "base_se_norm": [-0.6880443096160889, -0.28695839643478394], "gt_se_raw": [21.0, 30.0], "gt_se_norm": [-0.5320664644241333, -0.33152347803115845], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000216.pt"}
{"idx": 218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hgLQQe5uUCE.mp4", "duration": 89.75633333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip of the person getting on the bus and paying the fare.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip of the person getting on the bus and paying the fare.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 70.0], "base_se_norm": [0.09184491634368896, 0.5597785711288452], "gt_se_raw": [51.0, 70.0], "gt_se_norm": [0.13641011714935303, 0.5597785711288452], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000218.pt"}
{"idx": 220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VpPJP7o7NnA.mp4", "duration": 68.56, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the members of the Inclusion Team are presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the members of the Inclusion Team are presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 26.0], "base_se_norm": [-0.6499416828155518, -0.24154025316238403], "gt_se_raw": [16.0, 26.0], "gt_se_norm": [-0.5332555174827576, -0.24154025316238403], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000220.pt"}
{"idx": 222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VpPJP7o7NnA.mp4", "duration": 68.56, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment 'Universal Design for Learning' is highlighted as a goal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment 'Universal Design for Learning' is highlighted as a goal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [0.05017507076263428, 0.1668611764907837], "gt_se_raw": [36.0, 40.0], "gt_se_norm": [0.05017507076263428, 0.1668611764907837], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000222.pt"}
{"idx": 224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VpPJP7o7NnA.mp4", "duration": 68.56, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the self-destruct countdown sequence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the self-destruct countdown sequence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 59.0], "base_se_norm": [0.6044341325759888, 0.7211202383041382], "gt_se_raw": [55.0, 59.0], "gt_se_norm": [0.6044341325759888, 0.7211202383041382], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000224.pt"}
{"idx": 226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c8LraDxreeI.mp4", "duration": 129.56276666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is riding a chariot pulled by two white horses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is riding a chariot pulled by two white horses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.5832135677337646, -0.536903977394104], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.5986500978469849, -0.536903977394104], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000226.pt"}
{"idx": 228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c8LraDxreeI.mp4", "duration": 129.56276666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment where a woman in a grey dress is speaking into a microphone while a man holds an umbrella for her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment where a woman in a grey dress is speaking into a microphone while a man holds an umbrella for her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.6449596881866455, -0.5832135677337646], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [-0.6603962182998657, -0.5986500978469849], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000228.pt"}
{"idx": 230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uPUP8ey3MjE.mp4", "duration": 138.58333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the eagle attack another bird in the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the eagle attack another bird in the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 43.0], "base_se_norm": [-0.8412507772445679, -0.3794347047805786], "gt_se_raw": [25.0, 43.0], "gt_se_norm": [-0.6392062306404114, -0.3794347047805786], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000230.pt"}
{"idx": 232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uPUP8ey3MjE.mp4", "duration": 138.58333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the eagle attacks and pins down the wolf in the snow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the eagle attacks and pins down the wolf in the snow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 107.0], "base_se_norm": [0.3277209997177124, 0.5441973209381104], "gt_se_raw": [92.0, 98.0], "gt_se_norm": [0.3277209997177124, 0.4143115282058716], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000232.pt"}
{"idx": 234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uPUP8ey3MjE.mp4", "duration": 138.58333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a hunter on horseback releases an eagle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a hunter on horseback releases an eagle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 136.0], "base_se_norm": [0.8905593156814575, 0.9627180099487305], "gt_se_raw": [132.0, 135.0], "gt_se_norm": [0.9049910306930542, 0.9482862949371338], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000234.pt"}
{"idx": 236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TqPCGGHoxsE.mp4", "duration": 157.32383333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the two shoelaces are crossed for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the two shoelaces are crossed for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 87.0], "base_se_norm": [-0.03384000062942505, 0.10599899291992188], "gt_se_raw": [67.0, 72.0], "gt_se_norm": [-0.1482536792755127, -0.08469051122665405], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000236.pt"}
{"idx": 238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TqPCGGHoxsE.mp4", "duration": 157.32383333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two loops pulled to tighten the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two loops pulled to tighten the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 135.0], "base_se_norm": [0.6399295330047607, 0.7162052392959595], "gt_se_raw": [107.0, 111.0], "gt_se_norm": [0.3602515459060669, 0.41110217571258545], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000238.pt"}
{"idx": 240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cbMupML0UIo.mp4", "duration": 129.23333333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the logo for the Cerebral Palsy Alliance appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the logo for the Cerebral Palsy Alliance appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.8607170581817627, -0.7523858547210693], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.8607170581817627, -0.7523858547210693], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000240.pt"}
{"idx": 242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cbMupML0UIo.mp4", "duration": 129.23333333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video can we see the illustration of paper cut-out figures holding hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video can we see the illustration of paper cut-out figures holding hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.2107299566268921, -0.13335055112838745], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.2107299566268921, -0.13335055112838745], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000242.pt"}
{"idx": 244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cbMupML0UIo.mp4", "duration": 129.23333333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a silhouette illustration of children with various abilities is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a silhouette illustration of children with various abilities is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 109.0], "base_se_norm": [0.5166364908218384, 0.6868712902069092], "gt_se_raw": [98.0, 109.0], "gt_se_norm": [0.5166364908218384, 0.6868712902069092], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000244.pt"}
{"idx": 246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1AOXI0eBL8.mp4", "duration": 34.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when people are playing basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when people are playing basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.709864616394043, -0.5357834100723267], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.6518375277519226, -0.5938104391098022], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000246.pt"}
{"idx": 248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1AOXI0eBL8.mp4", "duration": 34.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman is serving food or drinks at an outdoor table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman is serving food or drinks at an outdoor table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.16054153442382812, 0.334622859954834], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.16054153442382812, 0.27659571170806885], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000248.pt"}
{"idx": 250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lwS74rI92YQ.mp4", "duration": 23.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is taking plates out of the cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is taking plates out of the cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.6605374813079834], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.8302687406539917, -0.6605374813079834], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000250.pt"}
{"idx": 252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lwS74rI92YQ.mp4", "duration": 23.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman is opening a drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman is opening a drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [0.2729843854904175, 0.5275812149047852], "gt_se_raw": [16.0, 17.0], "gt_se_norm": [0.3578500747680664, 0.4427156448364258], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000252.pt"}
{"idx": 254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lwS74rI92YQ.mp4", "duration": 23.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman sets the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman sets the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 23.0], "base_se_norm": [-0.6605374813079834, 0.9519094228744507], "gt_se_raw": [2.0, 22.0], "gt_se_norm": [-0.8302687406539917, 0.8670437335968018], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000254.pt"}
{"idx": 256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bf6D-i8YpHg.mp4", "duration": 189.08, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the dog makes a sad face in the coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the dog makes a sad face in the coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 62.0], "base_se_norm": [-0.3653479814529419, -0.3441929221153259], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [-0.3653479814529419, -0.3441929221153259], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000256.pt"}
{"idx": 258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bf6D-i8YpHg.mp4", "duration": 189.08, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the giant inflatable snowman runs over the Grinch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the giant inflatable snowman runs over the Grinch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 124.0], "base_se_norm": [0.2693040370941162, 0.31161415576934814], "gt_se_raw": [120.0, 122.0], "gt_se_norm": [0.2693040370941162, 0.2904590368270874], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000258.pt"}
{"idx": 260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0A8nEfYmmtM.mp4", "duration": 103.68691666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the playground equipment first appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the playground equipment first appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.5756455659866333, -0.5177791118621826], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.5563567876815796, -0.5177791118621826], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000260.pt"}
{"idx": 262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0A8nEfYmmtM.mp4", "duration": 103.68691666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a character says 'Let's just taco 'bout it'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a character says 'Let's just taco 'bout it'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.016269385814666748, 0.06088602542877197], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [0.0030194520950317383, 0.06088602542877197], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000262.pt"}
{"idx": 264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0A8nEfYmmtM.mp4", "duration": 103.68691666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the time when a crow is pecking at the dead chip at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when a crow is pecking at the dead chip at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 96.0], "base_se_norm": [0.6974176168441772, 0.8517282009124756], "gt_se_raw": [89.0, 98.0], "gt_se_norm": [0.716706395149231, 0.8903058767318726], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000264.pt"}
{"idx": 266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pCuE_A58ye0.mp4", "duration": 116.88, "prompt_text": "Please find the visual event described by the sentence 'At what point does Mr. Bean try to scare the ticket lady?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Mr. Bean try to scare the ticket lady?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 36.0], "base_se_norm": [-0.5722107887268066, -0.3839835524559021], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.5550992488861084, -0.5208761096000671], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000266.pt"}
{"idx": 268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pCuE_A58ye0.mp4", "duration": 116.88, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the cat in the movie transforms into a monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the cat in the movie transforms into a monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 64.0], "base_se_norm": [0.043805599212646484, 0.09514033794403076], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.06091725826263428, 0.09514033794403076], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000268.pt"}
{"idx": 270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IYnsfV5N2n8.mp4", "duration": 107.4, "prompt_text": "Please find the visual event described by the sentence 'When does the policeman shoot the gun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the policeman shoot the gun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.8882681727409363, -0.8510242104530334], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.8882681727409363, -0.8510242104530334], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000270.pt"}
{"idx": 272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IYnsfV5N2n8.mp4", "duration": 107.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with the 'T' on his chest turn into a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with the 'T' on his chest turn into a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [0.024208545684814453, 0.08007442951202393], "gt_se_raw": [56.0, 57.0], "gt_se_norm": [0.042830467224121094, 0.061452507972717285], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000272.pt"}
{"idx": 274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IYnsfV5N2n8.mp4", "duration": 107.4, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the cake is stabbed with a knife and starts screaming.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the cake is stabbed with a knife and starts screaming.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 73.0], "base_se_norm": [0.22905027866363525, 0.3594040870666504], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [0.2662942409515381, 0.3035382032394409], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000274.pt"}
{"idx": 276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7QORMPFSzpc.mp4", "duration": 270.27, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the chef pours beer into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the chef pours beer into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.6003996133804321, -0.5707995891571045], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.5929995775222778, -0.5707995891571045], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000276.pt"}
{"idx": 278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7QORMPFSzpc.mp4", "duration": 270.27, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a tortilla press is used to flatten the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a tortilla press is used to flatten the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 199.0], "base_se_norm": [0.2136012315750122, 0.4726015329360962], "gt_se_raw": [170.0, 172.0], "gt_se_norm": [0.25800132751464844, 0.2728012800216675], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000278.pt"}
{"idx": 280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmdAkgJDmEw.mp4", "duration": 194.31912499999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the YouTube logo on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the YouTube logo on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.6397678256034851, -0.6088907718658447], "gt_se_raw": [37.0, 38.0], "gt_se_norm": [-0.6191831231117249, -0.6088907718658447], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000280.pt"}
{"idx": 282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmdAkgJDmEw.mp4", "duration": 194.31912499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman folding a piece of paper and tucking it into her shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman folding a piece of paper and tucking it into her shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 94.0], "base_se_norm": [-0.15602749586105347, -0.03251928091049194], "gt_se_raw": [82.0, 92.0], "gt_se_norm": [-0.15602749586105347, -0.0531039834022522], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000282.pt"}
{"idx": 284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmdAkgJDmEw.mp4", "duration": 194.31912499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the man use the hammer to smash the candies on the table for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man use the hammer to smash the candies on the table for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 113.0], "base_se_norm": [0.10128116607666016, 0.16303527355194092], "gt_se_raw": [108.0, 109.0], "gt_se_norm": [0.11157357692718506, 0.12186586856842041], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000284.pt"}
{"idx": 286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahj8Vef9L24.mp4", "duration": 381.13075, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where people are playing ball against a graffiti wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where people are playing ball against a graffiti wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 65.0], "base_se_norm": [-0.7586129903793335, -0.6589096784591675], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.7533654570579529, -0.737622857093811], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000286.pt"}
{"idx": 288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahj8Vef9L24.mp4", "duration": 381.13075, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the police motorcade passes a roadblock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the police motorcade passes a roadblock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 83.0], "base_se_norm": [-0.6169293522834778, -0.5644539594650269], "gt_se_raw": [72.0, 77.0], "gt_se_norm": [-0.6221768856048584, -0.5959391593933105], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000288.pt"}
{"idx": 290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIpev8JXJHQ.mp4", "duration": 175.50866666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man is crawling through the dark tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man is crawling through the dark tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 51.0], "base_se_norm": [-0.623950183391571, -0.41883212327957153], "gt_se_raw": [31.0, 51.0], "gt_se_norm": [-0.6467410922050476, -0.41883212327957153], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000290.pt"}
{"idx": 292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIpev8JXJHQ.mp4", "duration": 175.50866666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman in a black dress sings on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman in a black dress sings on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 30.0], "base_se_norm": [-0.7948819398880005, -0.6581365466117859], "gt_se_raw": [99.0, 103.0], "gt_se_norm": [0.1281493902206421, 0.17373120784759521], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000292.pt"}
{"idx": 294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIpev8JXJHQ.mp4", "duration": 175.50866666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man yells in frustration, 'It's my head!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man yells in frustration, 'It's my head!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 162.0], "base_se_norm": [0.7662944793701172, 0.8460626602172852], "gt_se_raw": [155.0, 159.0], "gt_se_norm": [0.7662944793701172, 0.8118762969970703], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000294.pt"}
{"idx": 296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nCLk51ELofU.mp4", "duration": 47.347, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an adult's hand lowers the pom-pom onto the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an adult's hand lowers the pom-pom onto the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 24.0], "base_se_norm": [-0.28189748525619507, 0.013791799545288086], "gt_se_raw": [20.0, 27.0], "gt_se_norm": [-0.15517348051071167, 0.14051580429077148], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000296.pt"}
{"idx": 298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nCLk51ELofU.mp4", "duration": 47.347, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child's right hand touch the green tinsel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child's right hand touch the green tinsel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 21.0], "base_se_norm": [-0.3241388201713562, -0.11293220520019531], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [0.7318942546844482, 0.8586182594299316], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000298.pt"}
{"idx": 300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5t7CB2N9oyU.mp4", "duration": 60.10170833333333, "prompt_text": "Please find the visual event described by the sentence 'When is a woman having her hair brushed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman having her hair brushed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.23463070392608643, -0.16807687282562256], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.23463070392608643, -0.16807687282562256], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000300.pt"}
{"idx": 302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5t7CB2N9oyU.mp4", "duration": 60.10170833333333, "prompt_text": "Please find the visual event described by the sentence 'A man is sitting in a chair and getting a leg massage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is sitting in a chair and getting a leg massage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [-0.4342922568321228, -0.36773842573165894], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.4675692319869995, -0.40101534128189087], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000302.pt"}
{"idx": 304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5t7CB2N9oyU.mp4", "duration": 60.10170833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a woman in a black turtleneck is playing a drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a woman in a black turtleneck is playing a drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 49.0], "base_se_norm": [0.5640155076980591, 0.630569338798523], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.5640155076980591, 0.630569338798523], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000304.pt"}
{"idx": 306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tNeYTDLZUJA.mp4", "duration": 377.0099666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the rules are changed for letters, sending vowels to the right and consonants to the left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the rules are changed for letters, sending vowels to the right and consonants to the left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 51.0], "base_se_norm": [-0.819633424282074, -0.7294501066207886], "gt_se_raw": [29.0, 36.0], "gt_se_norm": [-0.8461579084396362, -0.8090236186981201], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000306.pt"}
{"idx": 308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tNeYTDLZUJA.mp4", "duration": 377.0099666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment describing the Fair Housing Act of 1968 and its rules against discrimination.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 217.0 - 229.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment describing the Fair Housing Act of 1968 and its rules against discrimination.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 217.0 - 229.0 seconds.<|im_end|>\n", "base_se_raw": [217.0, 229.0], "base_se_norm": [0.1511632204055786, 0.2148219347000122], "gt_se_raw": [217.0, 222.0], "gt_se_norm": [0.1511632204055786, 0.1776876449584961], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000308.pt"}
{"idx": 310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/chnJS5jGC5s.mp4", "duration": 123.15636666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman introduce herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman introduce herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.9512814283370972, -0.837604820728302], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.9188023805618286, -0.8700838685035706], "start_token_span": [61, 64], "end_token_span": [66, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000310.pt"}
{"idx": 312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/chnJS5jGC5s.mp4", "duration": 123.15636666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman points to and describes the bird's features like its eyes, ears, and mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman points to and describes the bird's features like its eyes, ears, and mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 72.0], "base_se_norm": [-0.4316168427467346, 0.16924524307250977], "gt_se_raw": [64.0, 72.0], "gt_se_norm": [0.0393291711807251, 0.16924524307250977], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000312.pt"}
{"idx": 314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/chnJS5jGC5s.mp4", "duration": 123.15636666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the woman stands up after finishing her story.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the woman stands up after finishing her story.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 122.0], "base_se_norm": [0.948742151260376, 0.9812211990356445], "gt_se_raw": [121.0, 123.0], "gt_se_norm": [0.9649816751480103, 0.9974607229232788], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000314.pt"}
{"idx": 316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MOQCYung8sA.mp4", "duration": 149.78296666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman interviewed in a store about the transport difficulties faced by the elderly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman interviewed in a store about the transport difficulties faced by the elderly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 141.0], "base_se_norm": [0.45543909072875977, 0.8827240467071533], "gt_se_raw": [109.0, 141.0], "gt_se_norm": [0.45543909072875977, 0.8827240467071533], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000316.pt"}
{"idx": 318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MOQCYung8sA.mp4", "duration": 149.78296666666668, "prompt_text": "Please find the visual event described by the sentence 'Where can we see an Opal card being used to tap on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see an Opal card being used to tap on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.4658939242362976, -0.4258359670639038], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.4658939242362976, -0.412483274936676], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000318.pt"}
{"idx": 320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5OLlemqd3SE.mp4", "duration": 70.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a person using a two-finger gesture to interact with the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a person using a two-finger gesture to interact with the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.6602170467376709, -0.5752713680267334], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [-0.6319018602371216, -0.6035865545272827], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000320.pt"}
{"idx": 322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5OLlemqd3SE.mp4", "duration": 70.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person controlling a yellow airplane in a video game with their hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person controlling a yellow airplane in a video game with their hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 51.0], "base_se_norm": [-0.15054267644882202, 0.4440774917602539], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.15054267644882202, -0.037281692028045654], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000322.pt"}
{"idx": 324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5OLlemqd3SE.mp4", "duration": 70.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a person uses both hands to interact with the data graphs on the monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a person uses both hands to interact with the data graphs on the monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 64.0], "base_se_norm": [0.6422841548919678, 0.8121756315231323], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.6705993413925171, 0.7838603258132935], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000324.pt"}
{"idx": 326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdpspllWI2o.mp4", "duration": 92.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the orangutan examines the shampoo bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the orangutan examines the shampoo bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.6739130020141602, -0.6086956262588501], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.6739130020141602, -0.6086956262588501], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000326.pt"}
{"idx": 328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdpspllWI2o.mp4", "duration": 92.0, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the girl is pointing at the orangutan, telling it to leave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the girl is pointing at the orangutan, telling it to leave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [-0.54347825050354, -0.47826087474823], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.54347825050354, -0.47826087474823], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000328.pt"}
{"idx": 330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1roy4o4tqQM.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man looking at a train ticket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man looking at a train ticket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.5771424770355225, -0.5207614302635193], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.5771424770355225, -0.5207614302635193], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000330.pt"}
{"idx": 332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1roy4o4tqQM.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a group of Bulbasaurs are walking in a creek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a group of Bulbasaurs are walking in a creek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 93.0], "base_se_norm": [0.282667875289917, 0.3108583688735962], "gt_se_raw": [91.0, 93.0], "gt_se_norm": [0.282667875289917, 0.3108583688735962], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000332.pt"}
{"idx": 334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1roy4o4tqQM.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the monsters are flying between the trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the monsters are flying between the trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 98.0], "base_se_norm": [0.35314416885375977, 0.38133466243743896], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.35314416885375977, 0.38133466243743896], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000334.pt"}
{"idx": 336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d61MkuYttDI.mp4", "duration": 192.05853333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does Elsa change Anna's dress with her magic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Elsa change Anna's dress with her magic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [-0.750076174736023, -0.7292492389678955], "gt_se_raw": [90.0, 92.0], "gt_se_norm": [-0.06278574466705322, -0.041958749294281006], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000336.pt"}
{"idx": 338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d61MkuYttDI.mp4", "duration": 192.05853333333334, "prompt_text": "Please find the visual event described by the sentence 'Show when a man and a reindeer are scared by a large snow monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show when a man and a reindeer are scared by a large snow monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 154.0], "base_se_norm": [0.5828508138656616, 0.6036777496337891], "gt_se_raw": [152.0, 154.0], "gt_se_norm": [0.5828508138656616, 0.6036777496337891], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000338.pt"}
{"idx": 340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQXRmsvjwNU.mp4", "duration": 84.83474444444445, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where all the ingredients for the meatballs are mixed together in a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where all the ingredients for the meatballs are mixed together in a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 28.0], "base_se_norm": [-0.5049198865890503, -0.339893102645874], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.5049198865890503, -0.3870435953140259], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000340.pt"}
{"idx": 342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQXRmsvjwNU.mp4", "duration": 84.83474444444445, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when coconut milk is poured into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when coconut milk is poured into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [-0.0805654525756836, -0.03341495990753174], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.0805654525756836, -0.03341495990753174], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000342.pt"}
{"idx": 344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQXRmsvjwNU.mp4", "duration": 84.83474444444445, "prompt_text": "Please find the visual event described by the sentence 'When can we see a fork cutting a meatball from the finished dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a fork cutting a meatball from the finished dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 66.0], "base_se_norm": [0.41451478004455566, 0.5559662580490112], "gt_se_raw": [60.0, 66.0], "gt_se_norm": [0.41451478004455566, 0.5559662580490112], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000344.pt"}
{"idx": 346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/84TouqfIsiI.mp4", "duration": 150.69220833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a man sprays a substance towards the back of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a man sprays a substance towards the back of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9601837396621704, -0.9070954322814941], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.7345582842826843, -0.7080141305923462], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000346.pt"}
{"idx": 348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/84TouqfIsiI.mp4", "duration": 150.69220833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the bald man is speaking while wearing a red turtleneck sweater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the bald man is speaking while wearing a red turtleneck sweater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 67.0], "base_se_norm": [-0.24349111318588257, -0.11077022552490234], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [-0.24349111318588257, -0.20367485284805298], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000348.pt"}
{"idx": 350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oTuTdMMPfvc.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the chocolate ice cream being placed into the coconut shell?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the chocolate ice cream being placed into the coconut shell?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9790143966674805, -0.8950721025466919], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9580288529396057, -0.9160577058792114], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000350.pt"}
{"idx": 352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oTuTdMMPfvc.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is someone pouring spiced turmeric honey sauce onto the ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is someone pouring spiced turmeric honey sauce onto the ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [-0.1186056137084961, -0.0766344666481018], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [-0.1186056137084961, -0.0766344666481018], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000352.pt"}
{"idx": 354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oTuTdMMPfvc.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'When are banana chips and other toppings added to the raw cacao base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are banana chips and other toppings added to the raw cacao base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 79.0], "base_se_norm": [0.40603387355804443, 0.6578608751296997], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [0.3850482702255249, 0.4689905643463135], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000354.pt"}
{"idx": 356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24f0OwnZE-Q.mp4", "duration": 192.15029166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man with the beard puts his foot on the log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man with the beard puts his foot on the log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.7606040835380554, -0.7293785214424133], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.7606040835380554, -0.7397870421409607], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000356.pt"}
{"idx": 358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24f0OwnZE-Q.mp4", "duration": 192.15029166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the plaid shirt list the tools he has in his shed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the plaid shirt list the tools he has in his shed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 113.0], "base_se_norm": [-0.2818121910095215, 0.17616260051727295], "gt_se_raw": [108.0, 113.0], "gt_se_norm": [0.12411999702453613, 0.17616260051727295], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000358.pt"}
{"idx": 360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gbicCoxSM1w.mp4", "duration": 66.2662, "prompt_text": "Please find the visual event described by the sentence 'When are the two circular green papers being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two circular green papers being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9094560742378235, -0.7585495710372925], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.8792747855186462, -0.7887309193611145], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000360.pt"}
{"idx": 362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gbicCoxSM1w.mp4", "duration": 66.2662, "prompt_text": "Please find the visual event described by the sentence 'A person guides another person's hands to assemble the paper pieces. When does this happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person guides another person's hands to assemble the paper pieces. When does this happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 52.0], "base_se_norm": [-0.6378244161605835, 0.5694276094436646], "gt_se_raw": [53.0, 61.0], "gt_se_norm": [0.5996088981628418, 0.8410593271255493], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000362.pt"}
{"idx": 364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gbicCoxSM1w.mp4", "duration": 66.2662, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the different paper pieces are assembled to form a larger structure and held up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the different paper pieces are assembled to form a larger structure and held up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 62.0], "base_se_norm": [0.5996088981628418, 0.8712406158447266], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [0.3581584692001343, 0.5694276094436646], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000364.pt"}
{"idx": 366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LbCBdiXpZNc.mp4", "duration": 88.95553333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a man is holding up a large, black egg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a man is holding up a large, black egg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [-0.6402697563171387, -0.5953034162521362], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.6402697563171387, -0.5953034162521362], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000366.pt"}
{"idx": 368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LbCBdiXpZNc.mp4", "duration": 88.95553333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a group of people are participating in a lecture at an art gallery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a group of people are participating in a lecture at an art gallery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.12315744161605835, -0.03322488069534302], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.12315744161605835, -0.07819116115570068], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000368.pt"}
{"idx": 370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MYWVkW1u-o4.mp4", "duration": 148.3, "prompt_text": "Please find the visual event described by the sentence 'Show me the woman sitting in a chair and speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the woman sitting in a chair and speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 25.0], "base_se_norm": [-0.8516520261764526, -0.6628456115722656], "gt_se_raw": [11.0, 25.0], "gt_se_norm": [-0.8516520261764526, -0.6628456115722656], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000370.pt"}
{"idx": 372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MYWVkW1u-o4.mp4", "duration": 148.3, "prompt_text": "Please find the visual event described by the sentence 'When are a man and a woman seen working on a mannequin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are a man and a woman seen working on a mannequin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 41.0], "base_se_norm": [-0.5549561977386475, -0.4470667839050293], "gt_se_raw": [33.0, 41.0], "gt_se_norm": [-0.5549561977386475, -0.4470667839050293], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000372.pt"}
{"idx": 374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MYWVkW1u-o4.mp4", "duration": 148.3, "prompt_text": "Please find the visual event described by the sentence 'At what point is a person holding up a white t-shirt with '1990' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a person holding up a white t-shirt with '1990' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 125.0], "base_se_norm": [0.5778826475143433, 0.6857720613479614], "gt_se_raw": [116.0, 125.0], "gt_se_norm": [0.5643965005874634, 0.6857720613479614], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000374.pt"}
{"idx": 376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4EqgGiJkrpU.mp4", "duration": 203.63676666666666, "prompt_text": "Please find the visual event described by the sentence 'What part of the video describes the work environment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What part of the video describes the work environment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 50.0], "base_se_norm": [-0.6857149004936218, -0.5089295506477356], "gt_se_raw": [33.0, 50.0], "gt_se_norm": [-0.6758934855461121, -0.5089295506477356], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000376.pt"}
{"idx": 378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4EqgGiJkrpU.mp4", "duration": 203.63676666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man outlines a typical workday.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man outlines a typical workday.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 99.0], "base_se_norm": [-0.2339300513267517, -0.02768045663833618], "gt_se_raw": [79.0, 99.0], "gt_se_norm": [-0.22410869598388672, -0.02768045663833618], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000378.pt"}
{"idx": 380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vYLaKMpqnOc.mp4", "duration": 67.3673, "prompt_text": "Please find the visual event described by the sentence 'When does the man start packing his bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start packing his bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [0.8703435659408569, 0.9890955686569214], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [0.8703435659408569, 0.9594075679779053], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000380.pt"}
{"idx": 382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vYLaKMpqnOc.mp4", "duration": 67.3673, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person in a grey top walks from right to left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person in a grey top walks from right to left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.8812479972839355, -0.7031201124191284], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8812479972839355, -0.8218720555305481], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000382.pt"}
{"idx": 384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vYLaKMpqnOc.mp4", "duration": 67.3673, "prompt_text": "Please find the visual event described by the sentence 'When is the man operating the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man operating the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 58.0], "base_se_norm": [-0.6734321117401123, 0.7219035625457764], "gt_se_raw": [10.0, 59.0], "gt_se_norm": [-0.7031201124191284, 0.7515915632247925], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000384.pt"}
{"idx": 386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-PN1gwDOa5I.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker talk about being a psychology major in college?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker talk about being a psychology major in college?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.535204291343689, -0.44224512577056885], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.40506142377853394, -0.34928596019744873], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000386.pt"}
{"idx": 388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-PN1gwDOa5I.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the speaker directly asks the audience for help and to consider making a donation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the speaker directly asks the audience for help and to consider making a donation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 91.0], "base_se_norm": [0.5803055763244629, 0.6918565034866333], "gt_se_raw": [82.0, 87.0], "gt_se_norm": [0.5245300531387329, 0.617489218711853], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000388.pt"}
{"idx": 390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elSiig58mOw.mp4", "duration": 156.78996666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the speaker demonstrate counting from one to four using the lower beads?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the speaker demonstrate counting from one to four using the lower beads?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 48.0], "base_se_norm": [-0.5025191307067871, -0.3877159357070923], "gt_se_raw": [41.0, 48.0], "gt_se_norm": [-0.47700732946395874, -0.3877159357070923], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000390.pt"}
{"idx": 392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elSiig58mOw.mp4", "duration": 156.78996666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman points out the column indicators on the abacus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman points out the column indicators on the abacus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 55.0], "base_se_norm": [-0.6045665144920349, -0.29842448234558105], "gt_se_raw": [104.0, 116.0], "gt_se_norm": [0.326615571975708, 0.4796864986419678], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000392.pt"}
{"idx": 394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6yd2tv4Ni4c.mp4", "duration": 88.2882, "prompt_text": "Please find the visual event described by the sentence 'When are the two women standing next to the black car and having a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women standing next to the black car and having a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 77.0], "base_se_norm": [-0.8867346048355103, 0.7442874908447266], "gt_se_raw": [5.0, 72.0], "gt_se_norm": [-0.8867346048355103, 0.6310219764709473], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000394.pt"}
{"idx": 396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6yd2tv4Ni4c.mp4", "duration": 88.2882, "prompt_text": "Please find the visual event described by the sentence 'At what point do the women discuss the event being 'virtual'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the women discuss the event being 'virtual'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 36.0], "base_se_norm": [-0.5922445058822632, -0.18448901176452637], "gt_se_raw": [19.0, 34.0], "gt_se_norm": [-0.5695914030075073, -0.2297951579093933], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000396.pt"}
{"idx": 398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6yd2tv4Ni4c.mp4", "duration": 88.2882, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the purple shirt pull the purple suitcase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the purple shirt pull the purple suitcase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 77.0], "base_se_norm": [-0.9093876481056213, 0.7442874908447266], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [0.6536750793457031, 0.7669405937194824], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000398.pt"}
{"idx": 400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oY6tCnu-1Do.mp4", "duration": 85.1851, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a character attempts to commit suicide with a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a character attempts to commit suicide with a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [-0.20173829793930054, -0.1547817587852478], "gt_se_raw": [35.0, 36.0], "gt_se_norm": [-0.17826002836227417, -0.1547817587852478], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000400.pt"}
{"idx": 402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oY6tCnu-1Do.mp4", "duration": 85.1851, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for when the person is punched in the face after being told there is something on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for when the person is punched in the face after being told there is something on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [-0.013912022113800049, 0.03304457664489746], "gt_se_raw": [43.0, 44.0], "gt_se_norm": [0.009566307067871094, 0.03304457664489746], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000402.pt"}
{"idx": 404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBQzXWMb7wk.mp4", "duration": 121.05426666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the robot mouse successfully reach the cheese for the final time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the robot mouse successfully reach the cheese for the final time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [0.8008452653884888, 0.8669313192367554], "gt_se_raw": [111.0, 113.0], "gt_se_norm": [0.8338882923126221, 0.8669313192367554], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000404.pt"}
{"idx": 406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBQzXWMb7wk.mp4", "duration": 121.05426666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the mouse make its first right turn in the initial demonstration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the mouse make its first right turn in the initial demonstration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9504354596138, -0.9008709192276001], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9504354596138, -0.9173924326896667], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000406.pt"}
{"idx": 408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBQzXWMb7wk.mp4", "duration": 121.05426666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the mouse activated to begin its final, successful run through the maze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the mouse activated to begin its final, successful run through the maze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 112.0], "base_se_norm": [0.42085039615631104, 0.850409746170044], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.4043288230895996, 0.45389342308044434], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000408.pt"}
{"idx": 410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x2-MCPa_3rU.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy's leg braces shatter as he runs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy's leg braces shatter as he runs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [-0.2205936312675476, -0.18347907066345215], "gt_se_raw": [59.0, 66.0], "gt_se_norm": [-0.27007973194122314, -0.18347907066345215], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000410.pt"}
{"idx": 412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x2-MCPa_3rU.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the three bullies pick up their bikes from the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the three bullies pick up their bikes from the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.8144270777702332, -0.7401978969573975], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.8020555377006531, -0.7649409770965576], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000412.pt"}
{"idx": 414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/94NanQuVkA4.mp4", "duration": 100.01657777777778, "prompt_text": "Please find the visual event described by the sentence 'When is the steak being sliced into strips?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the steak being sliced into strips?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9400099515914917, -0.8400264978408813], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.9200132489204407, -0.8400264978408813], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000414.pt"}
{"idx": 416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/94NanQuVkA4.mp4", "duration": 100.01657777777778, "prompt_text": "Please find the visual event described by the sentence 'At what time are the red bell peppers being cut and deseeded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the red bell peppers being cut and deseeded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [-0.12014585733413696, 0.019830942153930664], "gt_se_raw": [45.0, 51.0], "gt_se_norm": [-0.10014915466308594, 0.019830942153930664], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000416.pt"}
{"idx": 418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/94NanQuVkA4.mp4", "duration": 100.01657777777778, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where parsley is sprinkled over the baked peppers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where parsley is sprinkled over the baked peppers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 64.0], "base_se_norm": [0.23979449272155762, 0.2797877788543701], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.23979449272155762, 0.2797877788543701], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000418.pt"}
{"idx": 420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMBlu-8z_Mk.mp4", "duration": 142.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the stitches being cast onto the second needle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the stitches being cast onto the second needle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 68.0], "base_se_norm": [-0.44029849767684937, -0.0485074520111084], "gt_se_raw": [37.0, 57.0], "gt_se_norm": [-0.48227614164352417, -0.20242536067962646], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000420.pt"}
{"idx": 422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMBlu-8z_Mk.mp4", "duration": 142.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the process of knitting the very first stitch to join the work in the round.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of knitting the very first stitch to join the work in the round.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 68.0], "base_se_norm": [-0.44029849767684937, -0.0485074520111084], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.18936562538146973, 0.28731346130371094], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000422.pt"}
{"idx": 424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7HDtvRHyq2M.mp4", "duration": 114.23912499999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the students forming the university's mascot logo on the football field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the students forming the university's mascot logo on the football field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 80.0], "base_se_norm": [0.36555659770965576, 0.4005708694458008], "gt_se_raw": [78.0, 80.0], "gt_se_norm": [0.36555659770965576, 0.4005708694458008], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000424.pt"}
{"idx": 426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7HDtvRHyq2M.mp4", "duration": 114.23912499999999, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where people are helping to set up a blue tarp on a house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where people are helping to set up a blue tarp on a house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.6148430109024048, -0.5623216032981873], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.6148430109024048, -0.5623216032981873], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000426.pt"}
{"idx": 428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7HDtvRHyq2M.mp4", "duration": 114.23912499999999, "prompt_text": "Please find the visual event described by the sentence 'When are the group of students cheering and holding up foam fingers at a stadium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the group of students cheering and holding up foam fingers at a stadium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.8774500489234924, -0.8249286413192749], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8774500489234924, -0.8249286413192749], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000428.pt"}
{"idx": 430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ddh13c7Hdo8.mp4", "duration": 11.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment that discusses penalties for copyright infringement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment that discusses penalties for copyright infringement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.8188405632972717, 0.8115942478179932], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8115942478179932], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000430.pt"}
{"idx": 432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ddh13c7Hdo8.mp4", "duration": 11.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a message about unauthorized distribution is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a message about unauthorized distribution is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.8188405632972717, 0.8115942478179932], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8115942478179932], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000432.pt"}
{"idx": 434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDquUlVDdbU.mp4", "duration": 178.278, "prompt_text": "Please find the visual event described by the sentence 'When is the person scooping soil from the bin into a yogurt cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person scooping soil from the bin into a yogurt cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, -0.7531944513320923], "gt_se_raw": [0.0, 22.0], "gt_se_norm": [-1.0, -0.7531944513320923], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000434.pt"}
{"idx": 436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDquUlVDdbU.mp4", "duration": 178.278, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person is carrying the newly potted plants on a tray and walking through the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person is carrying the newly potted plants on a tray and walking through the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 168.0], "base_se_norm": [0.5144885778427124, 0.8846969604492188], "gt_se_raw": [99.0, 118.0], "gt_se_norm": [0.11062502861022949, 0.32377517223358154], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000436.pt"}
{"idx": 438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDquUlVDdbU.mp4", "duration": 178.278, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person become visibly frustrated, dropping and then kicking the cardboard tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person become visibly frustrated, dropping and then kicking the cardboard tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 162.0], "base_se_norm": [0.7388572692871094, 0.8173863887786865], "gt_se_raw": [158.0, 164.0], "gt_se_norm": [0.7725125551223755, 0.8398232460021973], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000438.pt"}
{"idx": 440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UNOKO0rFfgg.mp4", "duration": 141.72, "prompt_text": "Please find the visual event described by the sentence 'Show the moment from an overhead perspective where the shark is swimming among a school of fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment from an overhead perspective where the shark is swimming among a school of fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 41.0], "base_se_norm": [-0.5342929363250732, -0.4213942885398865], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.5484052896499634, -0.4213942885398865], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000440.pt"}
{"idx": 442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UNOKO0rFfgg.mp4", "duration": 141.72, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the close-up shot of the shark's underbelly and mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the close-up shot of the shark's underbelly and mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [-0.1814846396446228, -0.12503528594970703], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.1814846396446228, -0.13914763927459717], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000442.pt"}
{"idx": 444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9pH4c4fRgOg.mp4", "duration": 118.85206666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the man has an outburst after being touched.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the man has an outburst after being touched.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.7307577133178711, -0.6802748441696167], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.7307577133178711, -0.697102427482605], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000444.pt"}
{"idx": 446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9pH4c4fRgOg.mp4", "duration": 118.85206666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man is interacting with a hairbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man is interacting with a hairbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 19.0], "base_se_norm": [-0.9326894283294678, -0.6802748441696167], "gt_se_raw": [90.0, 96.0], "gt_se_norm": [0.5144877433776855, 0.6154536008834839], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000446.pt"}
{"idx": 448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9pH4c4fRgOg.mp4", "duration": 118.85206666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a woman presents a new set of items on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a woman presents a new set of items on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [0.2957284450531006, 0.3966943025588989], "gt_se_raw": [77.0, 80.0], "gt_se_norm": [0.2957284450531006, 0.346211314201355], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000448.pt"}
{"idx": 450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z3biFxZIJOQ.mp4", "duration": 148.8153333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the villain in the kabuki mask attacks with microbots in a dark warehouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the villain in the kabuki mask attacks with microbots in a dark warehouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 41.0], "base_se_norm": [-0.5430578589439392, -0.44898152351379395], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.5430578589439392, -0.502739429473877], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000450.pt"}
{"idx": 452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z3biFxZIJOQ.mp4", "duration": 148.8153333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment Baymax, with low battery, stumbles up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment Baymax, with low battery, stumbles up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 122.0], "base_se_norm": [0.5186606645584106, 0.6396158933639526], "gt_se_raw": [114.0, 116.0], "gt_se_norm": [0.5321000814437866, 0.5589790344238281], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000452.pt"}
{"idx": 454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B9hkIBvBxOQ.mp4", "duration": 304.6043, "prompt_text": "Please find the visual event described by the sentence 'When can we see the map of Alaska shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the map of Alaska shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 46.0], "base_se_norm": [-0.7701936960220337, -0.6979688405990601], "gt_se_raw": [36.0, 46.0], "gt_se_norm": [-0.7636277675628662, -0.6979688405990601], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000454.pt"}
{"idx": 456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B9hkIBvBxOQ.mp4", "duration": 304.6043, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man walk along a path towards the glacier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man walk along a path towards the glacier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 148.0], "base_se_norm": [-0.061076998710632324, -0.028247475624084473], "gt_se_raw": [144.0, 148.0], "gt_se_norm": [-0.054511070251464844, -0.028247475624084473], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000456.pt"}
{"idx": 458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B9hkIBvBxOQ.mp4", "duration": 304.6043, "prompt_text": "Please find the visual event described by the sentence 'When can we see the sea lions gathered on the rocks by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 244.0 - 258.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the sea lions gathered on the rocks by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 244.0 - 258.0 seconds.<|im_end|>\n", "base_se_raw": [244.0, 258.0], "base_se_norm": [0.6020784378051758, 0.6940009593963623], "gt_se_raw": [244.0, 249.0], "gt_se_norm": [0.6020784378051758, 0.6349079608917236], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000458.pt"}
{"idx": 460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bSEYSZLzEEo.mp4", "duration": 95.6956, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the 'blender drink' and 'massage' cards are visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the 'blender drink' and 'massage' cards are visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.8537027835845947, -0.7701044082641602], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.8537027835845947, -0.7701044082641602], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000460.pt"}
{"idx": 462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bSEYSZLzEEo.mp4", "duration": 95.6956, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man touches the 'blender drink' card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man touches the 'blender drink' card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.6447067260742188, -0.5611083507537842], "gt_se_raw": [76.0, 84.0], "gt_se_norm": [0.5883697271347046, 0.7555665969848633], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000462.pt"}
{"idx": 464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KLd5Q9IL_cU.mp4", "duration": 36.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see a white cane moving along the brick path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a white cane moving along the brick path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9444444179534912, -0.7222222089767456], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8333333134651184], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000464.pt"}
{"idx": 466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KLd5Q9IL_cU.mp4", "duration": 36.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a person holding a smartphone to record the scenery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a person holding a smartphone to record the scenery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.3888888955116272, -0.2222222089767456], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.3888888955116272, -0.2222222089767456], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000466.pt"}
{"idx": 468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KLd5Q9IL_cU.mp4", "duration": 36.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the text 'I'm more powerful than you think' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the text 'I'm more powerful than you think' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [0.5, 0.6666666269302368], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [0.4444444179534912, 0.6111111640930176], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000468.pt"}
{"idx": 470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OTmPw4iy0hk.mp4", "duration": 166.4663, "prompt_text": "Please find the visual event described by the sentence 'At what point is the UK startup SwiftKey and its predictive text app shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the UK startup SwiftKey and its predictive text app shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 75.0], "base_se_norm": [-0.1830177903175354, -0.09891670942306519], "gt_se_raw": [69.0, 75.0], "gt_se_norm": [-0.1710033416748047, -0.09891670942306519], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000470.pt"}
{"idx": 472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OTmPw4iy0hk.mp4", "duration": 166.4663, "prompt_text": "Please find the visual event described by the sentence 'Locate the animation illustrating how an infrared sensor detects cheek muscle movement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the animation illustrating how an infrared sensor detects cheek muscle movement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 136.0], "base_se_norm": [0.4777766466140747, 0.6339644193649292], "gt_se_raw": [106.0, 113.0], "gt_se_norm": [0.27353107929229736, 0.3576321601867676], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000472.pt"}
{"idx": 474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-vSmjbrUMpg.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'When are the rights related to each license discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the rights related to each license discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 64.0], "base_se_norm": [0.015384674072265625, 0.9692307710647583], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [0.16923081874847412, 0.3230769634246826], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000474.pt"}
{"idx": 476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-vSmjbrUMpg.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Where in the video is a summary of each of the six licenses mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video is a summary of each of the six licenses mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 64.0], "base_se_norm": [0.015384674072265625, 0.9692307710647583], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [0.6000000238418579, 0.7230769395828247], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000476.pt"}
{"idx": 478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-vSmjbrUMpg.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'When does the video display the list of licenses from least to most restrictive?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display the list of licenses from least to most restrictive?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 64.0], "base_se_norm": [-1.0, 0.9692307710647583], "gt_se_raw": [23.0, 37.0], "gt_se_norm": [-0.29230767488479614, 0.13846158981323242], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000478.pt"}
{"idx": 480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_0F7wFPRBmY.mp4", "duration": 188.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the hat pet her corgi?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the hat pet her corgi?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 145.0], "base_se_norm": [0.49655044078826904, 0.5390058755874634], "gt_se_raw": [140.0, 145.0], "gt_se_norm": [0.48593664169311523, 0.5390058755874634], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000480.pt"}
{"idx": 482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_0F7wFPRBmY.mp4", "duration": 188.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the puppy is playing with the bigger, yellow dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the puppy is playing with the bigger, yellow dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 39.0], "base_se_norm": [-0.7134264707565308, -0.5860605239868164], "gt_se_raw": [27.0, 39.0], "gt_se_norm": [-0.7134264707565308, -0.5860605239868164], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000482.pt"}
{"idx": 484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zi5ISjgijko.mp4", "duration": 76.81006666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a picture of two people at the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a picture of two people at the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 50.0], "base_se_norm": [0.1717214584350586, 0.30191266536712646], "gt_se_raw": [45.0, 54.0], "gt_se_norm": [0.1717214584350586, 0.4060657024383545], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000484.pt"}
{"idx": 486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zi5ISjgijko.mp4", "duration": 76.81006666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the brother explains how Jake communicates without the exact word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the brother explains how Jake communicates without the exact word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 65.0], "base_se_norm": [0.5102187395095825, 0.6924865245819092], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [0.5362570285797119, 0.6924865245819092], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000486.pt"}
{"idx": 488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zi5ISjgijko.mp4", "duration": 76.81006666666667, "prompt_text": "Please find the visual event described by the sentence 'Where is the selfie of the mother and son shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where is the selfie of the mother and son shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 70.0], "base_se_norm": [0.6924865245819092, 0.822677731513977], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [0.718524694442749, 0.8487160205841064], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000488.pt"}
{"idx": 490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QHVJvqC4LVE.mp4", "duration": 165.62379166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man brushing his teeth with a foamy mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man brushing his teeth with a foamy mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 85.0], "base_se_norm": [-0.2996175289154053, 0.026422619819641113], "gt_se_raw": [58.0, 85.0], "gt_se_norm": [-0.2996175289154053, 0.026422619819641113], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000490.pt"}
{"idx": 492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QHVJvqC4LVE.mp4", "duration": 165.62379166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment he starts chugging the orange juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment he starts chugging the orange juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 128.0], "base_se_norm": [0.41284048557281494, 0.5456715822219849], "gt_se_raw": [118.0, 128.0], "gt_se_norm": [0.42491602897644043, 0.5456715822219849], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000492.pt"}
{"idx": 494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQwD0QCbxaA.mp4", "duration": 55.0, "prompt_text": "Please find the visual event described by the sentence 'When does the hedgehog roll itself across the finish line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the hedgehog roll itself across the finish line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 36.0], "base_se_norm": [-0.12727272510528564, 0.30909085273742676], "gt_se_raw": [25.0, 26.0], "gt_se_norm": [-0.09090906381607056, -0.054545462131500244], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000494.pt"}
{"idx": 496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQwD0QCbxaA.mp4", "duration": 55.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does a bird place a bowl of leaves at the finish line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a bird place a bowl of leaves at the finish line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [0.27272725105285645, 0.38181817531585693], "gt_se_raw": [34.0, 35.0], "gt_se_norm": [0.23636364936828613, 0.27272725105285645], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000496.pt"}
{"idx": 498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQwD0QCbxaA.mp4", "duration": 55.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a stack of flies lands on the man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a stack of flies lands on the man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.6000000238418579, -0.418181836605072], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [-0.6000000238418579, -0.5636363625526428], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000498.pt"}
{"idx": 500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vp01GKtd2AQ.mp4", "duration": 58.47508333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the stores are closed and out of business.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the stores are closed and out of business.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.5553661584854126, -0.41855573654174805], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.5553661584854126, -0.41855573654174805], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000500.pt"}
{"idx": 502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vp01GKtd2AQ.mp4", "duration": 58.47508333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the dog licks the heart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the dog licks the heart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [0.026078104972839355, 0.1286858320236206], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.07652974128723145, -0.00812453031539917], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000502.pt"}
{"idx": 504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ri4xpRikZXg.mp4", "duration": 65.1651, "prompt_text": "Please find the visual event described by the sentence 'When does the boy with blonde hair throw his hands over his head in shock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy with blonde hair throw his hands over his head in shock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.6623959541320801, -0.5396308898925781], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.6623959541320801, -0.5703221559524536], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000504.pt"}
{"idx": 506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ri4xpRikZXg.mp4", "duration": 65.1651, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a boy opens a green can and gets scared by a spring snake popping out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a boy opens a green can and gets scared by a spring snake popping out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 48.0], "base_se_norm": [0.07419466972351074, 0.47318124771118164], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.35041606426239014, 0.4424898624420166], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000506.pt"}
{"idx": 508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ri4xpRikZXg.mp4", "duration": 65.1651, "prompt_text": "Please find the visual event described by the sentence 'At what point do they light the contents of the can on fire, causing an explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they light the contents of the can on fire, causing an explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 62.0], "base_se_norm": [0.6880201101303101, 0.902859091758728], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [0.8721678256988525, 0.9335503578186035], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000508.pt"}
{"idx": 510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSz0zs0v6e8.mp4", "duration": 68.26820000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point is there a close-up shot of the yellow warning sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there a close-up shot of the yellow warning sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.8828151226043701, -0.7949265241622925], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.8535189032554626, -0.8242226839065552], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000510.pt"}
{"idx": 512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSz0zs0v6e8.mp4", "duration": 68.26820000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the bald man approach the woman and touch her shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bald man approach the woman and touch her shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 38.0], "base_se_norm": [-0.09181731939315796, 0.11325621604919434], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.12111353874206543, -0.003928661346435547], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000512.pt"}
{"idx": 514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LSfJQkA-bKE.mp4", "duration": 81.91516666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the raspberry vinaigrette used to marinate the chicken?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the raspberry vinaigrette used to marinate the chicken?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 15.0], "base_se_norm": [-0.8535069823265076, -0.6337674856185913], "gt_se_raw": [6.0, 15.0], "gt_se_norm": [-0.8535069823265076, -0.6337674856185913], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000514.pt"}
{"idx": 516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LSfJQkA-bKE.mp4", "duration": 81.91516666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the cooked chicken being diced on the cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the cooked chicken being diced on the cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 38.0], "base_se_norm": [-0.24311941862106323, -0.07221090793609619], "gt_se_raw": [31.0, 38.0], "gt_se_norm": [-0.24311941862106323, -0.07221090793609619], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000516.pt"}
{"idx": 518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LSfJQkA-bKE.mp4", "duration": 81.91516666666666, "prompt_text": "Please find the visual event described by the sentence 'Where can I see the final dish being assembled into lettuce wraps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I see the final dish being assembled into lettuce wraps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 77.0], "base_se_norm": [0.5381766557693481, 0.8799936771392822], "gt_se_raw": [64.0, 73.0], "gt_se_norm": [0.5625921487808228, 0.7823317050933838], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000518.pt"}
{"idx": 520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpsh8dYl7PE.mp4", "duration": 40.33182707223309, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the animal turn to reveal a mask on its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the animal turn to reveal a mask on its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.45452505350112915, -0.25617051124572754], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.45452505350112915, -0.3057591915130615], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000520.pt"}
{"idx": 522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpsh8dYl7PE.mp4", "duration": 40.33182707223309, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the group of sheep runs away from the masked animal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the group of sheep runs away from the masked animal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 31.0], "base_se_norm": [0.19012713432312012, 0.5372475385665894], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [0.38848161697387695, 0.5372475385665894], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000522.pt"}
{"idx": 524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FUoGS9Nx05M.mp4", "duration": 150.4, "prompt_text": "Please find the visual event described by the sentence 'When can we see the exterior of the newly built University of Sydney accommodation building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the exterior of the newly built University of Sydney accommodation building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 14.0], "base_se_norm": [-0.9335106611251831, -0.813829779624939], "gt_se_raw": [6.0, 14.0], "gt_se_norm": [-0.9202127456665039, -0.813829779624939], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000524.pt"}
{"idx": 526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FUoGS9Nx05M.mp4", "duration": 150.4, "prompt_text": "Please find the visual event described by the sentence 'At what point is the stair platform lift for wheelchair users being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the stair platform lift for wheelchair users being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.375, -0.30851060152053833], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.3617020845413208, -0.30851060152053833], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000526.pt"}
{"idx": 528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FUoGS9Nx05M.mp4", "duration": 150.4, "prompt_text": "Please find the visual event described by the sentence 'When does the video demonstrate the adjustable height of the kitchen counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video demonstrate the adjustable height of the kitchen counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 86.0], "base_se_norm": [0.09042561054229736, 0.14361703395843506], "gt_se_raw": [83.0, 86.0], "gt_se_norm": [0.10372340679168701, 0.14361703395843506], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000528.pt"}
{"idx": 530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zXBmZLmfQZ4.mp4", "duration": 164.2, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the car drives into the zoo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the car drives into the zoo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 76.0], "base_se_norm": [-0.1230207085609436, -0.0742996335029602], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [-0.1230207085609436, -0.0986601710319519], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000530.pt"}
{"idx": 532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zXBmZLmfQZ4.mp4", "duration": 164.2, "prompt_text": "Please find the visual event described by the sentence 'When are the man, woman, and poodle falling towards the crocodile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man, woman, and poodle falling towards the crocodile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 83.0], "base_se_norm": [-0.049939095973968506, 0.010962247848510742], "gt_se_raw": [79.0, 82.0], "gt_se_norm": [-0.037758827209472656, -0.0012180209159851074], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000532.pt"}
{"idx": 534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/djqOMU5xpVo.mp4", "duration": 106.75055262710423, "prompt_text": "Please find the visual event described by the sentence 'When does Ganondorf attack Pikachu, causing a massive pink and yellow explosion against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Ganondorf attack Pikachu, causing a massive pink and yellow explosion against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 22.0], "base_se_norm": [-0.8126473426818848, -0.5878241062164307], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.8126473426818848, -0.7564415335655212], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000534.pt"}
{"idx": 536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/djqOMU5xpVo.mp4", "duration": 106.75055262710423, "prompt_text": "Please find the visual event described by the sentence 'At what point do many pink, biscuit-like objects fill the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do many pink, biscuit-like objects fill the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.5128830671310425, -0.40047144889831543], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.49414777755737305, -0.4379420280456543], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000536.pt"}
{"idx": 538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/djqOMU5xpVo.mp4", "duration": 106.75055262710423, "prompt_text": "Please find the visual event described by the sentence 'When does Roy use his fiery sword to attack Pichu on the icy ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Roy use his fiery sword to attack Pichu on the icy ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [0.4051450490951538, 0.4988213777542114], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.46135079860687256, 0.5175565481185913], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000538.pt"}
{"idx": 540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9rxfM1pHQrY.mp4", "duration": 138.37170503837172, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a model and a little girl are walking the runway together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a model and a little girl are walking the runway together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.7398312091827393, -0.6964697241783142], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.7398312091827393, -0.6964697241783142], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000540.pt"}
{"idx": 542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9rxfM1pHQrY.mp4", "duration": 138.37170503837172, "prompt_text": "Please find the visual event described by the sentence 'When are two women seen talking in a backstage or prep area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women seen talking in a backstage or prep area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.6097468137741089, -0.5519315004348755], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.5808391571044922, -0.5374777317047119], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000542.pt"}
{"idx": 544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UCuhC9jTFMM.mp4", "duration": 145.27012499999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an animated woman is thinking about a test paper with an 'A' grade.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an animated woman is thinking about a test paper with an 'A' grade.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [-0.6695810556411743, -0.6007437705993652], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.6420461535453796, -0.6007437705993652], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000544.pt"}
{"idx": 546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UCuhC9jTFMM.mp4", "duration": 145.27012499999998, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment featuring the pixelated game 'SCHOLARSHIP FINDER 2020'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment featuring the pixelated game 'SCHOLARSHIP FINDER 2020'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 56.0], "base_se_norm": [-0.4768366813659668, -0.22902250289916992], "gt_se_raw": [38.0, 56.0], "gt_se_norm": [-0.4768366813659668, -0.22902250289916992], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000546.pt"}
{"idx": 548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UCuhC9jTFMM.mp4", "duration": 145.27012499999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the words 'NO WAY WITHOUT' are displayed next to the woman in a yellow sweater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the words 'NO WAY WITHOUT' are displayed next to the woman in a yellow sweater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 82.0], "base_se_norm": [0.07386147975921631, 0.12893140316009521], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.07386147975921631, 0.11516392230987549], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000548.pt"}
{"idx": 550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbgMz9omXik.mp4", "duration": 20.353666666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman feeds the child a spoonful of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman feeds the child a spoonful of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [0.2774111032485962, 0.5721982717514038], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.2774111032485962, 0.47393596172332764], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000550.pt"}
{"idx": 552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbgMz9omXik.mp4", "duration": 20.353666666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the child has a bib on and is sitting at the table before eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the child has a bib on and is sitting at the table before eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.017376065254211426], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.8034752011299133, -0.41042560338974], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000552.pt"}
{"idx": 554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ObAw6Hv5n0k.mp4", "duration": 49.375, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the dog finding a bone on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the dog finding a bone on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.5139240622520447, -0.351898729801178], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.5139240622520447, -0.351898729801178], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000554.pt"}
{"idx": 556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ObAw6Hv5n0k.mp4", "duration": 49.375, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog bark and drop the bone into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog bark and drop the bone into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [0.3772151470184326, 0.45822787284851074], "gt_se_raw": [35.0, 39.0], "gt_se_norm": [0.4177215099334717, 0.5797468423843384], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000556.pt"}
{"idx": 558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ObAw6Hv5n0k.mp4", "duration": 49.375, "prompt_text": "Please find the visual event described by the sentence 'When is the moral of the story, 'Don't be greedy,' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the moral of the story, 'Don't be greedy,' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [0.7822785377502441, 0.9443037509918213], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [0.7822785377502441, 0.9443037509918213], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000558.pt"}
{"idx": 560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdkNEDjPes4.mp4", "duration": 147.31383333333332, "prompt_text": "Please find the visual event described by the sentence 'Where can we see several men playing drums on white buckets on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see several men playing drums on white buckets on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.8506589531898499, -0.7963531017303467], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.8506589531898499, -0.7963531017303467], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000560.pt"}
{"idx": 562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdkNEDjPes4.mp4", "duration": 147.31383333333332, "prompt_text": "Please find the visual event described by the sentence 'When does a black and white photo of a large group of people in front of a wooden building appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a black and white photo of a large group of people in front of a wooden building appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 63.0], "base_se_norm": [-0.17183607816696167, -0.1446831226348877], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.18541252613067627, -0.1446831226348877], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000562.pt"}
{"idx": 564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5E_SFw2V99E.mp4", "duration": 76.31596548725838, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'Lucille Ball' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'Lucille Ball' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.790345311164856, -0.7379316091537476], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.790345311164856, -0.7379316091537476], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000564.pt"}
{"idx": 566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5E_SFw2V99E.mp4", "duration": 76.31596548725838, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the name 'Thomas Edison' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the name 'Thomas Edison' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 42.0], "base_se_norm": [0.048273444175720215, 0.10068714618682861], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [0.048273444175720215, 0.10068714618682861], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000566.pt"}
{"idx": 568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5E_SFw2V99E.mp4", "duration": 76.31596548725838, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with the close-up of a spinning 'ONE CENT' coin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with the close-up of a spinning 'ONE CENT' coin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 57.0], "base_se_norm": [0.28413498401641846, 0.4937896728515625], "gt_se_raw": [49.0, 56.0], "gt_se_norm": [0.28413498401641846, 0.4675828218460083], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000568.pt"}
{"idx": 570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ofnCdC8P70g.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where Death and the deer are watching the full moon together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 178.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where Death and the deer are watching the full moon together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 178.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 178.0], "base_se_norm": [0.16666662693023682, 0.1866666078567505], "gt_se_raw": [175.0, 178.0], "gt_se_norm": [0.16666662693023682, 0.1866666078567505], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000570.pt"}
{"idx": 572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ofnCdC8P70g.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when it is snowing in the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 187.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when it is snowing in the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 187.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 187.0], "base_se_norm": [0.19333338737487793, 0.24666666984558105], "gt_se_raw": [179.0, 187.0], "gt_se_norm": [0.19333338737487793, 0.24666666984558105], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000572.pt"}
{"idx": 574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkjbMoj0JY4.mp4", "duration": 28.92, "prompt_text": "Please find the visual event described by the sentence 'When does the man take the first meal out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man take the first meal out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.861687421798706, -0.7233748435974121], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.861687421798706, -0.7233748435974121], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000574.pt"}
{"idx": 576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkjbMoj0JY4.mp4", "duration": 28.92, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man have a stack of four meals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man have a stack of four meals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 26.0], "base_se_norm": [0.10650074481964111, 0.7980636358261108], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.10650074481964111, 0.24481332302093506], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000576.pt"}
{"idx": 578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkjbMoj0JY4.mp4", "duration": 28.92, "prompt_text": "Please find the visual event described by the sentence 'When does the man say 'I will' and walk away from the food dispenser?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say 'I will' and walk away from the food dispenser?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.7289073467254639, 0.9363762140274048], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [0.7289073467254639, 0.9363762140274048], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000578.pt"}
{"idx": 580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CaS0mXRHfGc.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person is working on repairing a silver car's body.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person is working on repairing a silver car's body.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 121.0], "base_se_norm": [-0.06610244512557983, 0.08655393123626709], "gt_se_raw": [105.0, 121.0], "gt_se_norm": [-0.057122647762298584, 0.08655393123626709], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000580.pt"}
{"idx": 582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CaS0mXRHfGc.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'When is the sign for 'West Kentucky Community & Technical College' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the sign for 'West Kentucky Community & Technical College' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.640808641910553, -0.613869309425354], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.640808641910553, -0.613869309425354], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000582.pt"}
{"idx": 584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HV-AmXJsVyc.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the photo taken at Christmas at Wonderland in 2000.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the photo taken at Christmas at Wonderland in 2000.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.929315984249115], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.929315984249115], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000584.pt"}
{"idx": 586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HV-AmXJsVyc.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that defines an invisible disability through a Google search.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that defines an invisible disability through a Google search.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 61.0], "base_se_norm": [-0.010423541069030762, 0.07793152332305908], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.042589545249938965, 0.09560251235961914], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000586.pt"}
{"idx": 588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HV-AmXJsVyc.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the stick figure think 'Thank you for thinking of me'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the stick figure think 'Thank you for thinking of me'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 112.0], "base_se_norm": [0.9084688425064087, 0.9791529178619385], "gt_se_raw": [108.0, 112.0], "gt_se_norm": [0.9084688425064087, 0.9791529178619385], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000588.pt"}
{"idx": 590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b1XGPvbWn0A.mp4", "duration": 116.31620000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man in a gray jacket walks silently alongside the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man in a gray jacket walks silently alongside the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 69.0], "base_se_norm": [-0.1402745246887207, 0.18642115592956543], "gt_se_raw": [53.0, 69.0], "gt_se_norm": [-0.08869099617004395, 0.18642115592956543], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000590.pt"}
{"idx": 592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b1XGPvbWn0A.mp4", "duration": 116.31620000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman is walking through a crowded city square at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman is walking through a crowded city square at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 103.0], "base_se_norm": [0.6678674221038818, 0.7710344791412354], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.6678674221038818, 0.702256441116333], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000592.pt"}
{"idx": 594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y1jKP5m97Jk.mp4", "duration": 118.16666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket sit down at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket sit down at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.9153737425804138, -0.8476727604866028], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8815232515335083, -0.8476727604866028], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000594.pt"}
{"idx": 596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y1jKP5m97Jk.mp4", "duration": 118.16666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in black reveals he is autistic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in black reveals he is autistic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 63.0], "base_se_norm": [-0.17066287994384766, 0.06629061698913574], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [-0.05218613147735596, -0.01833564043045044], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000596.pt"}
{"idx": 598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y1jKP5m97Jk.mp4", "duration": 118.16666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket stand up and walk away from the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket stand up and walk away from the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [0.8448519706726074, 0.9125529527664185], "gt_se_raw": [110.0, 114.0], "gt_se_norm": [0.8617771863937378, 0.9294781684875488], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000598.pt"}
{"idx": 600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ei6TjvfGMpE.mp4", "duration": 101.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the flags of China and Australia are displayed together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the flags of China and Australia are displayed together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 29.0], "base_se_norm": [-0.606428325176239, -0.4293210506439209], "gt_se_raw": [20.0, 29.0], "gt_se_norm": [-0.606428325176239, -0.4293210506439209], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000600.pt"}
{"idx": 602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ei6TjvfGMpE.mp4", "duration": 101.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time range that explains what a 'Blind Sidewalk' is.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range that explains what a 'Blind Sidewalk' is.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 55.0], "base_se_norm": [-0.11446374654769897, 0.0823221206665039], "gt_se_raw": [46.0, 55.0], "gt_se_norm": [-0.09478515386581421, 0.0823221206665039], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000602.pt"}
{"idx": 604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M26ug8MGYlY.mp4", "duration": 139.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two moose fighting in the driveway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two moose fighting in the driveway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 35.0], "base_se_norm": [-1.0, -0.4983277916908264], "gt_se_raw": [0.0, 41.0], "gt_se_norm": [-1.0, -0.4123268127441406], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000604.pt"}
{"idx": 606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M26ug8MGYlY.mp4", "duration": 139.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the moose are fighting on the rocky ground next to the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the moose are fighting on the rocky ground next to the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 86.0], "base_se_norm": [-0.34065937995910645, 0.2326803207397461], "gt_se_raw": [59.0, 85.0], "gt_se_norm": [-0.1543239951133728, 0.21834683418273926], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000606.pt"}
{"idx": 608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M26ug8MGYlY.mp4", "duration": 139.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the losing moose run away, ending the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the losing moose run away, ending the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 128.0], "base_se_norm": [0.5766841173171997, 0.8346869945526123], "gt_se_raw": [122.0, 128.0], "gt_se_norm": [0.7486859560012817, 0.8346869945526123], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000608.pt"}
{"idx": 610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z2BgjH_CtIA.mp4", "duration": 23.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when the dog is actively running in its sleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the dog is actively running in its sleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 18.0], "base_se_norm": [-0.13669061660766602, 0.5539568662643433], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.05035966634750366, 0.5539568662643433], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000610.pt"}
{"idx": 612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z2BgjH_CtIA.mp4", "duration": 23.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the dog is standing up and looks confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the dog is standing up and looks confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 22.0], "base_se_norm": [0.726618766784668, 0.8992806673049927], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.726618766784668, 0.9856115579605103], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000612.pt"}
{"idx": 614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DCWHh_mZogE.mp4", "duration": 313.84, "prompt_text": "Please find the visual event described by the sentence 'When does Daddy Pig sit on his favorite sunny spot reading a newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Daddy Pig sit on his favorite sunny spot reading a newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 34.0], "base_se_norm": [-0.8789191842079163, -0.7833290696144104], "gt_se_raw": [20.0, 33.0], "gt_se_norm": [-0.8725465536117554, -0.7897017598152161], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000614.pt"}
{"idx": 616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DCWHh_mZogE.mp4", "duration": 313.84, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where birds are eating the seeds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where birds are eating the seeds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 128.0], "base_se_norm": [-0.24802446365356445, -0.18429774045944214], "gt_se_raw": [122.0, 131.0], "gt_se_norm": [-0.22253376245498657, -0.16517972946166992], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000616.pt"}
{"idx": 618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DCWHh_mZogE.mp4", "duration": 313.84, "prompt_text": "Please find the visual event described by the sentence 'When do all the pigs laugh and fall over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 282.0 - 285.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the pigs laugh and fall over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 282.0 - 285.0 seconds.<|im_end|>\n", "base_se_raw": [282.0, 285.0], "base_se_norm": [0.7970941066741943, 0.8162120580673218], "gt_se_raw": [281.0, 285.0], "gt_se_norm": [0.7907214164733887, 0.8162120580673218], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000618.pt"}
{"idx": 620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8PWlxfTFmyE.mp4", "duration": 50.2, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a car is shown parked on a sandwich.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a car is shown parked on a sandwich.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.4820716977119446, -0.3625497817993164], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.4820716977119446, -0.3625497817993164], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000620.pt"}
{"idx": 622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8PWlxfTFmyE.mp4", "duration": 50.2, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an astronaut lets go of the spaceship and floats away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an astronaut lets go of the spaceship and floats away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 37.0], "base_se_norm": [0.3545815944671631, 0.47410356998443604], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.3944222927093506, 0.47410356998443604], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000622.pt"}
{"idx": 624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RvzFTHdMPWw.mp4", "duration": 40.04004004004004, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the black hoodie speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the black hoodie speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, 0.19880008697509766], "gt_se_raw": [0.0, 24.0], "gt_se_norm": [-1.0, 0.19880008697509766], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000624.pt"}
{"idx": 626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RvzFTHdMPWw.mp4", "duration": 40.04004004004004, "prompt_text": "Please find the visual event described by the sentence 'At what point is the company's phone number visible on a blue background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the company's phone number visible on a blue background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 39.0], "base_se_norm": [0.2487499713897705, 0.9480500221252441], "gt_se_raw": [24.0, 40.0], "gt_se_norm": [0.19880008697509766, 0.9980000257492065], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000626.pt"}
{"idx": 628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RvzFTHdMPWw.mp4", "duration": 40.04004004004004, "prompt_text": "Please find the visual event described by the sentence 'When can you see the logo for JC Auto Collision?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can you see the logo for JC Auto Collision?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, 0.19880008697509766], "gt_se_raw": [0.0, 40.0], "gt_se_norm": [-1.0, 0.9980000257492065], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000628.pt"}
{"idx": 630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-AXQVPtK-E.mp4", "duration": 188.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the player in the yellow shirt is standing next to a music box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the player in the yellow shirt is standing next to a music box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.734089732170105, -0.6915440559387207], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.7553625106811523, -0.734089732170105], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000630.pt"}
{"idx": 632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-AXQVPtK-E.mp4", "duration": 188.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the word 'VICTORY' displayed on the screen for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the word 'VICTORY' displayed on the screen for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 105.0], "base_se_norm": [0.09555041790008545, 0.11682319641113281], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [0.17000532150268555, 0.1912781000137329], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000632.pt"}
{"idx": 634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yad0gj-ZDuE.mp4", "duration": 119.9, "prompt_text": "Please find the visual event described by the sentence 'When can we see all the individual colored hearts laid out before they are assembled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see all the individual colored hearts laid out before they are assembled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 64.0], "base_se_norm": [-0.015846550464630127, 0.06755626201629639], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.015846550464630127, 0.05087566375732422], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000634.pt"}
{"idx": 636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yad0gj-ZDuE.mp4", "duration": 119.9, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when glue is being applied to a paper unit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when glue is being applied to a paper unit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 67.0], "base_se_norm": [0.06755626201629639, 0.11759793758392334], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [0.06755626201629639, 0.11759793758392334], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000636.pt"}
{"idx": 638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yad0gj-ZDuE.mp4", "duration": 119.9, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where multiple paper hearts are being joined together to form a ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where multiple paper hearts are being joined together to form a ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 95.0], "base_se_norm": [0.06755626201629639, 0.5846538543701172], "gt_se_raw": [75.0, 94.0], "gt_se_norm": [0.2510424852371216, 0.567973256111145], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000638.pt"}
{"idx": 640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NooW_RbfdWI.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man gently touch the chin of the baby dinosaur?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man gently touch the chin of the baby dinosaur?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.6432451009750366, -0.5802884101867676], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.6432451009750366, -0.5802884101867676], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000640.pt"}
{"idx": 642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NooW_RbfdWI.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the gyrosphere is rolling on the grass while being chased by dinosaurs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the gyrosphere is rolling on the grass while being chased by dinosaurs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 50.0], "base_se_norm": [-0.055648863315582275, 0.049278974533081055], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.055648863315582275, -0.013677716255187988], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000642.pt"}
{"idx": 644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pq6qrbKiDIk.mp4", "duration": 154.22073333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker make a finger gun gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker make a finger gun gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 39.0], "base_se_norm": [-0.5201683044433594, -0.4942314028739929], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [-0.5720419883728027, -0.546105146408081], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000644.pt"}
{"idx": 646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pq6qrbKiDIk.mp4", "duration": 154.22073333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker transition to discussing how people can be a problem?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker transition to discussing how people can be a problem?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 146.0], "base_se_norm": [0.7507373094558716, 0.8933900594711304], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [-0.001431286334991455, 0.024505615234375], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000646.pt"}
{"idx": 648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pq6qrbKiDIk.mp4", "duration": 154.22073333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the speaker mentions population pyramids.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the speaker mentions population pyramids.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 135.0], "base_se_norm": [0.6729267835617065, 0.7507373094558716], "gt_se_raw": [130.0, 136.0], "gt_se_norm": [0.6858952045440674, 0.7637057304382324], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000648.pt"}
{"idx": 650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U5oHhI_GmJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman with blonde hair is editing a video on a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman with blonde hair is editing a video on a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5414421558380127, -0.08288431167602539], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000650.pt"}
{"idx": 652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U5oHhI_GmJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a man in a black t-shirt smiles at the camera with his arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a man in a black t-shirt smiles at the camera with his arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3756735324859619, 0.8342313766479492], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000652.pt"}
{"idx": 654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DOyum_nw7NU.mp4", "duration": 220.95406666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a group of people are posing for a photo on a lawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a group of people are posing for a photo on a lawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 37.0], "base_se_norm": [-0.728450357913971, -0.6650887727737427], "gt_se_raw": [30.0, 37.0], "gt_se_norm": [-0.728450357913971, -0.6650887727737427], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000654.pt"}
{"idx": 656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DOyum_nw7NU.mp4", "duration": 220.95406666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where students are shown sitting in a classroom and listening attentively.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 178.0 - 202.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where students are shown sitting in a classroom and listening attentively.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 178.0 - 202.0 seconds.<|im_end|>\n", "base_se_raw": [178.0, 202.0], "base_se_norm": [0.6111946105957031, 0.8284343481063843], "gt_se_raw": [178.0, 188.0], "gt_se_norm": [0.6111946105957031, 0.7017111778259277], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000656.pt"}
{"idx": 658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DOyum_nw7NU.mp4", "duration": 220.95406666666668, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the close-up shot of the professional counselor license?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the close-up shot of the professional counselor license?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 68.0], "base_se_norm": [-0.42069411277770996, -0.38448745012283325], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [-0.4116424322128296, -0.3935391306877136], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000658.pt"}
{"idx": 660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5yGNbyAmkVY.mp4", "duration": 18.666666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the character with a TV head is walking on the sidewalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the character with a TV head is walking on the sidewalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 12.0], "base_se_norm": [-1.0, 0.28571438789367676], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.17857146263122559], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000660.pt"}
{"idx": 662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5yGNbyAmkVY.mp4", "duration": 18.666666666666668, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the character is standing next to the caution sign, just before falling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the character is standing next to the caution sign, just before falling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.03571426868438721, 0.28571438789367676], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.1428571343421936, 0.17857146263122559], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000662.pt"}
{"idx": 664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jwPc0kK9VHU.mp4", "duration": 123.5, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man sitting under a tree before an apple falls on him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man sitting under a tree before an apple falls on him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.7570850253105164, -0.6437246799468994], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.7570850253105164, -0.659919023513794], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000664.pt"}
{"idx": 666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jwPc0kK9VHU.mp4", "duration": 123.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man drops two apples to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man drops two apples to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.20647776126861572, -0.17408907413482666], "gt_se_raw": [50.0, 51.0], "gt_se_norm": [-0.1902834177017212, -0.17408907413482666], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000666.pt"}
{"idx": 668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jwPc0kK9VHU.mp4", "duration": 123.5, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man jumps off a building, attempting to fly with wings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man jumps off a building, attempting to fly with wings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 109.0], "base_se_norm": [0.6194331645965576, 0.7651821374893188], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.7327935695648193, 0.7651821374893188], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000668.pt"}
{"idx": 670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8zkVKHy1hyM.mp4", "duration": 115.0732888888889, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of kneading the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of kneading the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 41.0], "base_se_norm": [-0.4438326954841614, -0.2874106168746948], "gt_se_raw": [33.0, 41.0], "gt_se_norm": [-0.4264524579048157, -0.2874106168746948], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000670.pt"}
{"idx": 672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8zkVKHy1hyM.mp4", "duration": 115.0732888888889, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when mozzarella cheese is put on the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when mozzarella cheese is put on the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 68.0], "base_se_norm": [0.14709508419036865, 0.18185555934906006], "gt_se_raw": [66.0, 67.0], "gt_se_norm": [0.14709508419036865, 0.16447532176971436], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000672.pt"}
{"idx": 674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/60kjpwyQhqg.mp4", "duration": 15.708333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the camera zoom in on the character's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera zoom in on the character's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.6180371046066284, -0.36339521408081055], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.6180371046066284, -0.4907161593437195], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000674.pt"}
{"idx": 676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/60kjpwyQhqg.mp4", "duration": 15.708333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'Just in time' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'Just in time' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [0.6551724672317505, 0.9098143577575684], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.6551724672317505, 0.9098143577575684], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000676.pt"}
{"idx": 678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/60kjpwyQhqg.mp4", "duration": 15.708333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the character standing with its arms open to the sides?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the character standing with its arms open to the sides?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.8726790547370911, -0.10875332355499268], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.018567681312561035], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000678.pt"}
{"idx": 680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EXJ4Dbdm1ks.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the movie 'Moonlight' is mentioned as an example of intersectional media.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the movie 'Moonlight' is mentioned as an example of intersectional media.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 102.0], "base_se_norm": [-0.06158357858657837, 0.05185139179229736], "gt_se_raw": [92.0, 102.0], "gt_se_norm": [-0.05127131938934326, 0.05185139179229736], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000680.pt"}
{"idx": 682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EXJ4Dbdm1ks.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the video discusses non-intersectional queer representation in media like 'Call Me By Your Name' and 'Brokeback Mountain'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the video discusses non-intersectional queer representation in media like 'Call Me By Your Name' and 'Brokeback Mountain'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 91.0], "base_se_norm": [-0.1750185489654541, -0.06158357858657837], "gt_se_raw": [80.0, 91.0], "gt_se_norm": [-0.1750185489654541, -0.06158357858657837], "start_token_span": [86, 90], "end_token_span": [92, 96], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000682.pt"}
{"idx": 684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkbsMObgMBE.mp4", "duration": 127.96, "prompt_text": "Please find the visual event described by the sentence 'When can we see the snow leopard on the mountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the snow leopard on the mountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 59.0], "base_se_norm": [-0.10909658670425415, -0.07783681154251099], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [-0.10909658670425415, -0.07783681154251099], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000684.pt"}
{"idx": 686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkbsMObgMBE.mp4", "duration": 127.96, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a black panther is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a black panther is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 96.0], "base_se_norm": [0.3910597562789917, 0.5004688501358032], "gt_se_raw": [90.0, 96.0], "gt_se_norm": [0.4066896438598633, 0.5004688501358032], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000686.pt"}
{"idx": 688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkbsMObgMBE.mp4", "duration": 127.96, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the leopard cub is on a tree branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the leopard cub is on a tree branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 69.0], "base_se_norm": [0.00031256675720214844, 0.07846200466156006], "gt_se_raw": [65.0, 69.0], "gt_se_norm": [0.01594245433807373, 0.07846200466156006], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000688.pt"}
{"idx": 690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vCTNioSW730.mp4", "duration": 174.44, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the squirrel gets caught in an avalanche and slides down the mountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the squirrel gets caught in an avalanche and slides down the mountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 138.0], "base_se_norm": [0.28410911560058594, 0.5822058916091919], "gt_se_raw": [111.0, 120.0], "gt_se_norm": [0.27264392375946045, 0.3758312463760376], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000690.pt"}
{"idx": 692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vCTNioSW730.mp4", "duration": 174.44, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the squirrel is falling through the sky after being launched off the cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the squirrel is falling through the sky after being launched off the cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 157.0], "base_se_norm": [0.6624627113342285, 0.8000458478927612], "gt_se_raw": [146.0, 152.0], "gt_se_norm": [0.6739280223846436, 0.7427195310592651], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000692.pt"}
{"idx": 694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fPh8BTdzXzg.mp4", "duration": 309.9763333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man start blending the mango pieces, coconut milk, and syrup together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start blending the mango pieces, coconut milk, and syrup together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 113.0], "base_se_norm": [-0.3225288987159729, -0.2709120512008667], "gt_se_raw": [105.0, 119.0], "gt_se_norm": [-0.3225288987159729, -0.23219943046569824], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000694.pt"}
{"idx": 696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fPh8BTdzXzg.mp4", "duration": 309.9763333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the mango slices being placed into the blender?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the mango slices being placed into the blender?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 68.0], "base_se_norm": [-0.6322299838066101, -0.5612567663192749], "gt_se_raw": [58.0, 68.0], "gt_se_norm": [-0.6257778406143188, -0.5612567663192749], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000696.pt"}
{"idx": 698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fPh8BTdzXzg.mp4", "duration": 309.9763333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the sorbet is being scooped into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 281.0 - 286.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the sorbet is being scooped into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 281.0 - 286.0 seconds.<|im_end|>\n", "base_se_raw": [281.0, 286.0], "base_se_norm": [0.8130416870117188, 0.8453022241592407], "gt_se_raw": [281.0, 286.0], "gt_se_norm": [0.8130416870117188, 0.8453022241592407], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000698.pt"}
{"idx": 700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GbycvPwr1Wg.mp4", "duration": 147.93945833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the purple sweater knock over the glass of milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the purple sweater knock over the glass of milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [0.09504258632659912, 0.13559973239898682], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.09504258632659912, 0.13559973239898682], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000700.pt"}
{"idx": 702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GbycvPwr1Wg.mp4", "duration": 147.93945833333333, "prompt_text": "Please find the visual event described by the sentence 'A man is on his knees, eating something off the kitchen floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is on his knees, eating something off the kitchen floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 122.0], "base_se_norm": [0.5817282199859619, 0.6493233442306519], "gt_se_raw": [117.0, 122.0], "gt_se_norm": [0.5817282199859619, 0.6493233442306519], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000702.pt"}
{"idx": 704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y1TceAcocv0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator state that the employment rate for Americans without disabilities is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator state that the employment rate for Americans without disabilities is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9031662940979004, -0.4189978241920471], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9031662940979004, -0.4189978241920471], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000704.pt"}
{"idx": 706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y1TceAcocv0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker begin their general observation about the employment trend?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker begin their general observation about the employment trend?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.03166300058364868], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.3221641182899475, -0.1284967064857483], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000706.pt"}
{"idx": 708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T_FdjXqSZlc.mp4", "duration": 266.89996666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the father take the batteries out of the toy rocket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the father take the batteries out of the toy rocket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 32.0], "base_se_norm": [-0.8726114630699158, -0.7602097988128662], "gt_se_raw": [18.0, 26.0], "gt_se_norm": [-0.8651180267333984, -0.8051704168319702], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000708.pt"}
{"idx": 710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T_FdjXqSZlc.mp4", "duration": 266.89996666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the family seen walking on the sandy path through the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the family seen walking on the sandy path through the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 151.0], "base_se_norm": [0.08654940128326416, 0.13151013851165771], "gt_se_raw": [159.0, 171.0], "gt_se_norm": [0.1914576292037964, 0.28137898445129395], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000710.pt"}
{"idx": 712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T_FdjXqSZlc.mp4", "duration": 266.89996666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boy walks out of the store and onto the empty street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boy walks out of the store and onto the empty street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 122.0], "base_se_norm": [-0.1457473635673523, -0.08579981327056885], "gt_se_raw": [115.0, 119.0], "gt_se_norm": [-0.13825392723083496, -0.10828012228012085], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000712.pt"}
{"idx": 714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C9O28ne6bG8.mp4", "duration": 111.3112, "prompt_text": "Please find the visual event described by the sentence 'When does the orange cat play with the lobster's antenna?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the orange cat play with the lobster's antenna?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 74.0], "base_se_norm": [0.22179973125457764, 0.3296055793762207], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [0.22179973125457764, 0.3116379976272583], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000714.pt"}
{"idx": 716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C9O28ne6bG8.mp4", "duration": 111.3112, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a cat jumps over the bowl, causing a splash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a cat jumps over the bowl, causing a splash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.06009101867675781, 0.13196158409118652], "gt_se_raw": [42.0, 43.0], "gt_se_norm": [-0.24535900354385376, -0.22739136219024658], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000716.pt"}
{"idx": 718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tbBzXKN32Sk.mp4", "duration": 472.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the 'BADLANDSCHUGS' logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'BADLANDSCHUGS' logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 21.0], "base_se_norm": [-0.9322512149810791, -0.9110797643661499], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.9237826466560364, -0.9153140783309937], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000718.pt"}
{"idx": 720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tbBzXKN32Sk.mp4", "duration": 472.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when all three glasses are full and arranged on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when all three glasses are full and arranged on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 162.0], "base_se_norm": [-0.4114326238632202, -0.3140437602996826], "gt_se_raw": [165.0, 193.0], "gt_se_norm": [-0.30134087800979614, -0.18278056383132935], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000720.pt"}
{"idx": 722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tbBzXKN32Sk.mp4", "duration": 472.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man chugs the second glass of orange liquid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 235.0 - 271.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man chugs the second glass of orange liquid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 235.0 - 271.0 seconds.<|im_end|>\n", "base_se_raw": [235.0, 271.0], "base_se_norm": [-0.004940032958984375, 0.14749467372894287], "gt_se_raw": [242.0, 271.0], "gt_se_norm": [0.024700045585632324, 0.14749467372894287], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000722.pt"}
{"idx": 724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/adzYW5DZoWs.mp4", "duration": 123.83204166666665, "prompt_text": "Please find the visual event described by the sentence 'A person runs with a blue lightsaber in hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person runs with a blue lightsaber in hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 65.0], "base_se_norm": [-0.11169999837875366, 0.04980909824371338], "gt_se_raw": [60.0, 64.0], "gt_se_norm": [-0.03094547986984253, 0.03365814685821533], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000724.pt"}
{"idx": 726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/adzYW5DZoWs.mp4", "duration": 123.83204166666665, "prompt_text": "Please find the visual event described by the sentence 'Two women are seen hugging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Two women are seen hugging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 96.0], "base_se_norm": [0.5020345449447632, 0.5504872798919678], "gt_se_raw": [93.0, 96.0], "gt_se_norm": [0.5020345449447632, 0.5504872798919678], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000726.pt"}
{"idx": 728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YzN4Zj9S6so.mp4", "duration": 157.62413333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video display a pie chart about people with disabilities in the UK?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display a pie chart about people with disabilities in the UK?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 32.0], "base_se_norm": [-0.8223622441291809, -0.593970775604248], "gt_se_raw": [13.0, 32.0], "gt_se_norm": [-0.835050642490387, -0.593970775604248], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000728.pt"}
{"idx": 730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YzN4Zj9S6so.mp4", "duration": 157.62413333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the video compare narrow and wide supermarket aisles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the video compare narrow and wide supermarket aisles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 89.0], "base_se_norm": [-0.23869526386260986, 0.12926876544952393], "gt_se_raw": [59.0, 89.0], "gt_se_norm": [-0.2513836622238159, 0.12926876544952393], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000730.pt"}
{"idx": 732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YzN4Zj9S6so.mp4", "duration": 157.62413333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range for the final summary slide titled 'Design for dignity'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range for the final summary slide titled 'Design for dignity'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 157.0], "base_se_norm": [0.7383124828338623, 0.992080807685852], "gt_se_raw": [137.0, 157.0], "gt_se_norm": [0.7383124828338623, 0.992080807685852], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000732.pt"}
{"idx": 734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1t7VOZCyuJ0.mp4", "duration": 73.72, "prompt_text": "Please find the visual event described by the sentence 'When is the camera's display screen shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the camera's display screen shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 11.0], "base_se_norm": [-0.8643516302108765, -0.7015734910964966], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8914812803268433, -0.8372219204902649], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000734.pt"}
{"idx": 736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1t7VOZCyuJ0.mp4", "duration": 73.72, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman is in a wheelchair and the man has a prosthetic leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman is in a wheelchair and the man has a prosthetic leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 36.0], "base_se_norm": [-0.4302767515182495, -0.023331522941589355], "gt_se_raw": [21.0, 31.0], "gt_se_norm": [-0.4302767515182495, -0.15897995233535767], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000736.pt"}
{"idx": 738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WdC4DaYIeQ.mp4", "duration": 110.48537499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the snowman sneeze, causing his carrot nose to fly away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the snowman sneeze, causing his carrot nose to fly away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.6741650104522705, -0.6379611492156982], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.6922669410705566, -0.6198592185974121], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000738.pt"}
{"idx": 740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WdC4DaYIeQ.mp4", "duration": 110.48537499999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the reindeer gets its tongue stuck on the frozen lake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the reindeer gets its tongue stuck on the frozen lake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.601757287979126, -0.547451376914978], "gt_se_raw": [57.0, 58.0], "gt_se_norm": [0.031810760498046875, 0.04991269111633301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000740.pt"}
{"idx": 742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WdC4DaYIeQ.mp4", "duration": 110.48537499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the snowman and reindeer are chaotically slipping and sliding on the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the snowman and reindeer are chaotically slipping and sliding on the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 63.0], "base_se_norm": [-0.38453394174575806, 0.14042246341705322], "gt_se_raw": [34.0, 51.0], "gt_se_norm": [-0.38453394174575806, -0.0768008828163147], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000742.pt"}
{"idx": 744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_RCAWNlJZCE.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the conversation between the father and the daughter take place?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the conversation between the father and the daughter take place?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 33.0], "base_se_norm": [-0.0907968282699585, 0.11124837398529053], "gt_se_raw": [15.0, 24.0], "gt_se_norm": [-0.494887113571167, -0.191819429397583], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000744.pt"}
{"idx": 746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_RCAWNlJZCE.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment is a woman in a green jacket talking on a landline phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is a woman in a green jacket talking on a landline phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.0907968282699585, 0.14492249488830566], "gt_se_raw": [28.0, 35.0], "gt_se_norm": [-0.057122647762298584, 0.17859673500061035], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000746.pt"}
{"idx": 748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elXXLAiktZc.mp4", "duration": 87.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the woman and the girl reading a book together on the orange couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman and the girl reading a book together on the orange couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 86.0], "base_se_norm": [-1.0, 0.9762543439865112], "gt_se_raw": [0.0, 86.0], "gt_se_norm": [-1.0, 0.9762543439865112], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000748.pt"}
{"idx": 750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elXXLAiktZc.mp4", "duration": 87.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman makes the sign for the number 'three'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman makes the sign for the number 'three'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.655304491519928, -0.5404059886932373], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.6323248147964478, -0.5404059886932373], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000750.pt"}
{"idx": 752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elXXLAiktZc.mp4", "duration": 87.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman making the sign for 'love'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman making the sign for 'love'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9770203232765198, -0.8621218204498291], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [0.47070086002349854, 0.5166603326797485], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000752.pt"}
{"idx": 754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9ZTX302mNw4.mp4", "duration": 19.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the son sings 'I thought today was a lot of fun'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the son sings 'I thought today was a lot of fun'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.8979591727256775, -0.18367350101470947], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.3877550959587097, -0.18367350101470947], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000754.pt"}
{"idx": 756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9ZTX302mNw4.mp4", "duration": 19.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the text 'Introducing Time Machine' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the text 'Introducing Time Machine' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.632652997970581, 0.8367346525192261], "gt_se_raw": [16.0, 17.0], "gt_se_norm": [0.632652997970581, 0.7346938848495483], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000756.pt"}
{"idx": 758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9mmmipGP1h0.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where two men in plaid shirts are working on the internal components of a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where two men in plaid shirts are working on the internal components of a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.5737595558166504, -0.2540792226791382], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.5737595558166504, -0.2540792226791382], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000758.pt"}
{"idx": 760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9mmmipGP1h0.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of hands adjusting the knobs on a set of pressure gauges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of hands adjusting the knobs on a set of pressure gauges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.17216122150421143, 0.49184155464172363], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.17216122150421143, 0.49184155464172363], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000760.pt"}
{"idx": 762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KYazqIHYqNI.mp4", "duration": 195.46193333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see the puppy being bottle-fed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the puppy being bottle-fed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 88.0], "base_se_norm": [-0.19165843725204468, -0.09956890344619751], "gt_se_raw": [80.0, 88.0], "gt_se_norm": [-0.1814262866973877, -0.09956890344619751], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000762.pt"}
{"idx": 764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KYazqIHYqNI.mp4", "duration": 195.46193333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the puppy struggles to walk with wobbly back legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the puppy struggles to walk with wobbly back legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 125.0], "base_se_norm": [0.03344929218292236, 0.2790215015411377], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.24832499027252197, 0.2790215015411377], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000764.pt"}
{"idx": 766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KYazqIHYqNI.mp4", "duration": 195.46193333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the grown-up dog is licking the woman's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 180.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the grown-up dog is licking the woman's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 180.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [180.0, 183.0], "base_se_norm": [0.8417909145355225, 0.8724874258041382], "gt_se_raw": [180.0, 183.0], "gt_se_norm": [0.8417909145355225, 0.8724874258041382], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000766.pt"}
{"idx": 768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h8u_mj2IOcs.mp4", "duration": 54.854800000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher smile and lean towards the student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher smile and lean towards the student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 23.0], "base_se_norm": [-0.5989412069320679, -0.16142255067825317], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.19788241386413574, -0.01558297872543335], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000768.pt"}
{"idx": 770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h8u_mj2IOcs.mp4", "duration": 54.854800000000004, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the teacher and student are both looking at the newly turned page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the teacher and student are both looking at the newly turned page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [0.23963630199432373, 0.4583956003189087], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [0.3490159511566162, 0.5677752494812012], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000770.pt"}
{"idx": 772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C97ZSYzwabo.mp4", "duration": 36.236000000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the woman first place the green dinosaur in the tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman first place the green dinosaur in the tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9448062777519226, -0.6688376069068909], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9448062777519226, -0.6688376069068909], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000772.pt"}
{"idx": 774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C97ZSYzwabo.mp4", "duration": 36.236000000000004, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the child first touches the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the child first touches the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 16.0], "base_se_norm": [-0.6136438846588135, -0.11690032482147217], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.6136438846588135, -0.5032564401626587], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000774.pt"}
{"idx": 776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C97ZSYzwabo.mp4", "duration": 36.236000000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the child touch the green dinosaur while the woman is holding it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child touch the green dinosaur while the woman is holding it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 25.0], "base_se_norm": [0.15906834602355957, 0.37984323501586914], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [0.4350370168685913, 0.5454244613647461], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000776.pt"}
{"idx": 778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmvB4oKrF1c.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man is standing and playing the guitar in a studio, with a view into the control room through a window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man is standing and playing the guitar in a studio, with a view into the control room through a window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.5737595558166504, -0.2540792226791382], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.5737595558166504, -0.2540792226791382], "start_token_span": [81, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000778.pt"}
{"idx": 780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmvB4oKrF1c.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Find the part where someone is adjusting a sound mixing board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where someone is adjusting a sound mixing board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.17216122150421143, 0.49184155464172363], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.17216122150421143, 0.49184155464172363], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000780.pt"}
{"idx": 782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fUybrJmFv0c.mp4", "duration": 184.75123333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a butterfly with orange, black and white spots on a white flower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a butterfly with orange, black and white spots on a white flower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 29.0], "base_se_norm": [-0.8051433563232422, -0.6860643625259399], "gt_se_raw": [19.0, 29.0], "gt_se_norm": [-0.7943180203437805, -0.6860643625259399], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000782.pt"}
{"idx": 784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fUybrJmFv0c.mp4", "duration": 184.75123333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see the complete life cycle diagram of a butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see the complete life cycle diagram of a butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 76.0], "base_se_norm": [-0.26387500762939453, -0.17727208137512207], "gt_se_raw": [67.0, 76.0], "gt_se_norm": [-0.274700403213501, -0.17727208137512207], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000784.pt"}
{"idx": 786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fUybrJmFv0c.mp4", "duration": 184.75123333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the butterfly emerge from its chrysalis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the butterfly emerge from its chrysalis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 117.0], "base_se_norm": [0.17996501922607422, 0.2665679454803467], "gt_se_raw": [110.0, 117.0], "gt_se_norm": [0.19079041481018066, 0.2665679454803467], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000786.pt"}
{"idx": 788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XAyP11gQ22w.mp4", "duration": 171.43793333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in the black top pats the other woman's shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in the black top pats the other woman's shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 114.0], "base_se_norm": [0.3065953254699707, 0.3299274444580078], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.5916889905929565, -0.5683568716049194], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000788.pt"}
{"idx": 790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XAyP11gQ22w.mp4", "duration": 171.43793333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman in the green hoodie gestures with both hands open, palms facing up, while talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman in the green hoodie gestures with both hands open, palms facing up, while talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.9066717624664307, -0.8483415842056274], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.8950057029724121, -0.8716737031936646], "start_token_span": [78, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000790.pt"}
{"idx": 792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L0SPXFTfVRY.mp4", "duration": 36.302933333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy rubbing his eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy rubbing his eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, 0.9833108186721802], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [0.046747446060180664, 0.2671152353286743], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000792.pt"}
{"idx": 794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L0SPXFTfVRY.mp4", "duration": 36.302933333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman is bending over and holding the boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman is bending over and holding the boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, 0.9833108186721802], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9449080228805542, -0.66944819688797], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000794.pt"}
{"idx": 796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L0SPXFTfVRY.mp4", "duration": 36.302933333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the part where the boy's right hand is exploring the contents of the yellow bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the part where the boy's right hand is exploring the contents of the yellow bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 36.0], "base_se_norm": [0.4874831438064575, 0.9833108186721802], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.28380441665649414, -0.008344590663909912], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000796.pt"}
{"idx": 798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0g9_bCAb9AM.mp4", "duration": 154.38756666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a woman reading a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a woman reading a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 99.0], "base_se_norm": [0.21771454811096191, 0.2824866771697998], "gt_se_raw": [94.0, 100.0], "gt_se_norm": [0.21771454811096191, 0.29544103145599365], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000798.pt"}
{"idx": 800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0g9_bCAb9AM.mp4", "duration": 154.38756666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a phone displaying a navigation map shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a phone displaying a navigation map shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 116.0], "base_se_norm": [0.45089399814605713, 0.5027116537094116], "gt_se_raw": [111.0, 116.0], "gt_se_norm": [0.43793952465057373, 0.5027116537094116], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000800.pt"}
{"idx": 802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_MC3XuMvsDI.mp4", "duration": 130.38025, "prompt_text": "Please find the visual event described by the sentence 'When are the mother's emotions shown sitting at the control panel for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the mother's emotions shown sitting at the control panel for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.8159230351448059, -0.6932050585746765], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.8159230351448059, -0.7545640468597412], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000802.pt"}
{"idx": 804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_MC3XuMvsDI.mp4", "duration": 130.38025, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the little girl rolls her eyes in frustration.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the little girl rolls her eyes in frustration.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 65.0], "base_se_norm": [-0.0335959792137146, -0.0029164552688598633], "gt_se_raw": [64.0, 65.0], "gt_se_norm": [-0.018256187438964844, -0.0029164552688598633], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000804.pt"}
{"idx": 806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_MC3XuMvsDI.mp4", "duration": 130.38025, "prompt_text": "Please find the visual event described by the sentence 'Show me when the father angrily says 'That's it, go to your room!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the father angrily says 'That's it, go to your room!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 103.0], "base_se_norm": [0.5493143796920776, 0.5799939632415771], "gt_se_raw": [102.0, 104.0], "gt_se_norm": [0.5646541118621826, 0.5953336954116821], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000806.pt"}
{"idx": 808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/04pcSyzwoTg.mp4", "duration": 89.04, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where 'suspensions,' 'expulsions,' and 'in-school arrests' are mentioned as outcomes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where 'suspensions,' 'expulsions,' and 'in-school arrests' are mentioned as outcomes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.6630727648735046, -0.5732254981994629], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.6630727648735046, -0.5732254981994629], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000808.pt"}
{"idx": 810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/04pcSyzwoTg.mp4", "duration": 89.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a comparison of expulsion rates between black students and their white classmates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a comparison of expulsion rates between black students and their white classmates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [0.23539984226226807, 0.3027852773666382], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [0.23539984226226807, 0.3027852773666382], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000810.pt"}
{"idx": 812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sdUUx5FdySs.mp4", "duration": 189.55603333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the kiwi tying a rope around the base of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the kiwi tying a rope around the base of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.7151237726211548, -0.6729199290275574], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.7151237726211548, -0.6729199290275574], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000812.pt"}
{"idx": 814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sdUUx5FdySs.mp4", "duration": 189.55603333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the kiwi places trees onto the cliffside it built.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the kiwi places trees onto the cliffside it built.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 82.0], "base_se_norm": [-0.21922820806503296, -0.13482046127319336], "gt_se_raw": [74.0, 82.0], "gt_se_norm": [-0.21922820806503296, -0.13482046127319336], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000814.pt"}
{"idx": 816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sdUUx5FdySs.mp4", "duration": 189.55603333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me when the kiwi is flying through the air past the trees it placed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the kiwi is flying through the air past the trees it placed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 148.0], "base_se_norm": [0.17115771770477295, 0.5615435838699341], "gt_se_raw": [110.0, 148.0], "gt_se_norm": [0.1606067419052124, 0.5615435838699341], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000816.pt"}
{"idx": 818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJcE1mqSJDA.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman in the yellow shirt smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman in the yellow shirt smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.7000914216041565, -0.6001219153404236], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.7334145903587341, -0.6001219153404236], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000818.pt"}
{"idx": 820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJcE1mqSJDA.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where people are standing at the voting booths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where people are standing at the voting booths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 45.0], "base_se_norm": [0.2662806510925293, 0.49954283237457275], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [0.23295748233795166, 0.33292698860168457], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000820.pt"}
{"idx": 822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oOOCx287v4Q.mp4", "duration": 50.84245833333333, "prompt_text": "Please find the visual event described by the sentence 'A boy points to the alphabet learning card on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A boy points to the alphabet learning card on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9606627821922302, -0.6853023767471313], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.8819884061813354, -0.7246395945549011], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000822.pt"}
{"idx": 824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oOOCx287v4Q.mp4", "duration": 50.84245833333333, "prompt_text": "Please find the visual event described by the sentence 'A woman is teaching a child how to use an oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is teaching a child how to use an oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 23.0], "base_se_norm": [-0.3706047534942627, -0.09524434804916382], "gt_se_raw": [16.0, 23.0], "gt_se_norm": [-0.3706047534942627, -0.09524434804916382], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000824.pt"}
{"idx": 826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oOOCx287v4Q.mp4", "duration": 50.84245833333333, "prompt_text": "Please find the visual event described by the sentence 'Two children in car seats hold hands with each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Two children in car seats hold hands with each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 49.0], "base_se_norm": [0.6521624326705933, 0.9275228977203369], "gt_se_raw": [42.0, 48.0], "gt_se_norm": [0.6521624326705933, 0.8881857395172119], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000826.pt"}
{"idx": 828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9uP7WjixVFM.mp4", "duration": 119.49437499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man begin to chug the drink from the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man begin to chug the drink from the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 74.0], "base_se_norm": [-0.3305124044418335, 0.23855197429656982], "gt_se_raw": [40.0, 74.0], "gt_se_norm": [-0.3305124044418335, 0.23855197429656982], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000828.pt"}
{"idx": 830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9uP7WjixVFM.mp4", "duration": 119.49437499999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man lets out a loud burp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man lets out a loud burp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [0.23855197429656982, 0.30550074577331543], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.30550074577331543, 0.35571229457855225], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000830.pt"}
{"idx": 832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BXx51SR_3Sk.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the basketball hits the blue bucket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the basketball hits the blue bucket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.8786706924438477, -0.8517085909843445], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.8921517133712769, -0.8651896119117737], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000832.pt"}
{"idx": 834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BXx51SR_3Sk.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the green jacket face off against the man in the blue jersey while holding the basketball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the green jacket face off against the man in the blue jersey while holding the basketball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 74.0], "base_se_norm": [-0.35291022062301636, -0.00240325927734375], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.35291022062301636, -0.3259481191635132], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000834.pt"}
{"idx": 836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BXx51SR_3Sk.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man slam dunk the basketball, causing the backboard to shatter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man slam dunk the basketball, causing the backboard to shatter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 115.0], "base_se_norm": [0.4424710273742676, 0.5503193140029907], "gt_se_raw": [109.0, 114.0], "gt_se_norm": [0.469433069229126, 0.5368382930755615], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000836.pt"}
{"idx": 838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IAq8pEFNeJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man is inspecting the tire of a large truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man is inspecting the tire of a large truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5414421558380127, -0.08288431167602539], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000838.pt"}
{"idx": 840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IAq8pEFNeJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'YOU'VE GOT THIS!' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'YOU'VE GOT THIS!' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5066900253295898, 0.8342313766479492], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.5066900253295898, 0.8342313766479492], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000840.pt"}
{"idx": 842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgQVj4iMm8Y.mp4", "duration": 184.88488488488488, "prompt_text": "Please find the visual event described by the sentence 'When does Chris Evans appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Chris Evans appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.6322035789489746, -0.5672982931137085], "gt_se_raw": [34.0, 40.0], "gt_se_norm": [-0.6322035789489746, -0.5672982931137085], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000842.pt"}
{"idx": 844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgQVj4iMm8Y.mp4", "duration": 184.88488488488488, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where a man is eating a green apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where a man is eating a green apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 61.0], "base_se_norm": [-0.3617650270462036, -0.3401299715042114], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [-0.3617650270462036, -0.3401299715042114], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000844.pt"}
{"idx": 846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgQVj4iMm8Y.mp4", "duration": 184.88488488488488, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the title 'MEAN TWEETS 10' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the title 'MEAN TWEETS 10' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9891824722290039, -0.9567298293113708], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9783649444580078, -0.9567298293113708], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000846.pt"}
{"idx": 848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1awM6kmpd2g.mp4", "duration": 144.6445, "prompt_text": "Please find the visual event described by the sentence 'Find the shot that shows the wheelchair accessible space on the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot that shows the wheelchair accessible space on the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 32.0], "base_se_norm": [-0.6543248891830444, -0.5575358867645264], "gt_se_raw": [25.0, 32.0], "gt_se_norm": [-0.6543248891830444, -0.5575358867645264], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000848.pt"}
{"idx": 850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1awM6kmpd2g.mp4", "duration": 144.6445, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a high-angle view of the construction area on the platform.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a high-angle view of the construction area on the platform.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 95.0], "base_se_norm": [0.2444303035736084, 0.31356537342071533], "gt_se_raw": [90.0, 95.0], "gt_se_norm": [0.2444303035736084, 0.31356537342071533], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000850.pt"}
{"idx": 852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jUPGpJ4NRkg.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'When is the girl drawing a blue and green pattern on a white cap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl drawing a blue and green pattern on a white cap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.9333999156951904, -0.8667998909950256], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9333999156951904, -0.8667998909950256], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000852.pt"}
{"idx": 854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jUPGpJ4NRkg.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'At what point is a large robot riding a unicycle visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a large robot riding a unicycle visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.1341991424560547, -0.0009990334510803223], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.16749918460845947, -0.06759911775588989], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000854.pt"}
{"idx": 856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jUPGpJ4NRkg.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman is dancing inside her newly opened store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman is dancing inside her newly opened store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [0.5318014621734619, 0.7316017150878906], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.5318014621734619, 0.5984015464782715], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000856.pt"}
{"idx": 858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9Byx6TxOPx4.mp4", "duration": 59.7597, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the teacher takes the student's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the teacher takes the student's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.8995978832244873, -0.7991957664489746], "gt_se_raw": [24.0, 25.0], "gt_se_norm": [-0.1967831254005432, -0.1633157730102539], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000858.pt"}
{"idx": 860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9Byx6TxOPx4.mp4", "duration": 59.7597, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the words 'a big car' are being shown on the monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the words 'a big car' are being shown on the monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 52.0], "base_se_norm": [-0.4645220637321472, 0.7403032779693604], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.6064337491989136, 0.7403032779693604], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000860.pt"}
{"idx": 862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dXyrxjIcWdw.mp4", "duration": 77.49408333333332, "prompt_text": "Please find the visual event described by the sentence 'A man in a trench coat walks up to a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a trench coat walks up to a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 23.0], "base_se_norm": [-0.7677241563796997, -0.4064062833786011], "gt_se_raw": [9.0, 23.0], "gt_se_norm": [-0.7677241563796997, -0.4064062833786011], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000862.pt"}
{"idx": 864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dXyrxjIcWdw.mp4", "duration": 77.49408333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man attempts to turn the doorknob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man attempts to turn the doorknob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 36.0], "base_se_norm": [-0.25155574083328247, -0.07089674472808838], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [-0.22574728727340698, -0.09670519828796387], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000864.pt"}
{"idx": 866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dXyrxjIcWdw.mp4", "duration": 77.49408333333332, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man inspects the area around the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man inspects the area around the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 26.0], "base_se_norm": [-0.6902989149093628, -0.3289809823036194], "gt_se_raw": [10.0, 19.0], "gt_se_norm": [-0.741915762424469, -0.5096399784088135], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000866.pt"}
{"idx": 868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ff5aCoq6xn8.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the page for counting '1 Sock'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the page for counting '1 Sock'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 28.0], "base_se_norm": [-0.5371423959732056, -0.07428479194641113], "gt_se_raw": [14.0, 27.0], "gt_se_norm": [-0.5371423959732056, -0.1073460578918457], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000868.pt"}
{"idx": 870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ff5aCoq6xn8.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the time frame where three shells are presented for counting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time frame where three shells are presented for counting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 48.0], "base_se_norm": [0.256327748298645, 0.5869404077529907], "gt_se_raw": [39.0, 48.0], "gt_se_norm": [0.2893890142440796, 0.5869404077529907], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000870.pt"}
{"idx": 872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wDNxuom6Ge4.mp4", "duration": 68.4684, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher showing the card for 'avocado'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher showing the card for 'avocado'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 27.0], "base_se_norm": [-0.5326310992240906, -0.21131497621536255], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.9707894325256348, -0.8539472222328186], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000872.pt"}
{"idx": 874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wDNxuom6Ge4.mp4", "duration": 68.4684, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the student in the red shirt is using the tablet to input numbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the student in the red shirt is using the tablet to input numbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 53.0], "base_se_norm": [-0.03605163097381592, 0.5481594800949097], "gt_se_raw": [43.0, 53.0], "gt_se_norm": [0.2560539245605469, 0.5481594800949097], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000874.pt"}
{"idx": 876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wDNxuom6Ge4.mp4", "duration": 68.4684, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a student covers her face with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a student covers her face with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 15.0], "base_se_norm": [-0.9415788650512695, -0.5618416666984558], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.8986861705780029, 0.9863177537918091], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000876.pt"}
{"idx": 878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1g4AUYiz0LU.mp4", "duration": 16.541666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the striped shirt clap her hands and say 'Yay' to praise the child?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the striped shirt clap her hands and say 'Yay' to praise the child?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.329974889755249, 0.5717884302139282], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.758186399936676, -0.2745591402053833], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000878.pt"}
{"idx": 880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1g4AUYiz0LU.mp4", "duration": 16.541666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the child looks up and makes a vocal sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the child looks up and makes a vocal sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.032745540142059326, 0.20906805992126465], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [0.20906805992126465, 0.329974889755249], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000880.pt"}
{"idx": 882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nfosQcd4l5Q.mp4", "duration": 18.585233333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see a boy in a red and black striped shirt playing in a container of water and soil?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a boy in a red and black striped shirt playing in a container of water and soil?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 18.0], "base_se_norm": [-1.0, 0.9370216131210327], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, 0.5065723657608032], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000882.pt"}
{"idx": 884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nfosQcd4l5Q.mp4", "duration": 18.585233333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy raises his hand, showing it's covered in mud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy raises his hand, showing it's covered in mud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.35432612895965576, -0.13910150527954102], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.35432612895965576, -0.13910150527954102], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000884.pt"}
{"idx": 886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nfosQcd4l5Q.mp4", "duration": 18.585233333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the woman in the grey top reaches towards the flowers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the woman in the grey top reaches towards the flowers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.721796989440918, 0.9370216131210327], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.721796989440918, 0.9370216131210327], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000886.pt"}
{"idx": 888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tloBMf_KmX4.mp4", "duration": 175.24, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man is walking through the station gates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man is walking through the station gates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 64.0], "base_se_norm": [-0.34946364164352417, -0.2695731520652771], "gt_se_raw": [56.0, 62.0], "gt_se_norm": [-0.36087656021118164, -0.29239898920059204], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000888.pt"}
{"idx": 890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tloBMf_KmX4.mp4", "duration": 175.24, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the staff is preparing the ramp for the wheelchair to board the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the staff is preparing the ramp for the wheelchair to board the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 96.0], "base_se_norm": [0.004336833953857422, 0.09564018249511719], "gt_se_raw": [92.0, 97.0], "gt_se_norm": [0.049988508224487305, 0.10705316066741943], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000890.pt"}
{"idx": 892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LrI4FmRIHpI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a group of people are talking in a harvested field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a group of people are talking in a harvested field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000892.pt"}
{"idx": 894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LrI4FmRIHpI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a group of people are standing next to a large green combine harvester.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a group of people are standing next to a large green combine harvester.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08188092708587646, 0.3771786689758301], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000894.pt"}
{"idx": 896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LrI4FmRIHpI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'LENGTH of PROGRAM' displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'LENGTH of PROGRAM' displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000896.pt"}
{"idx": 898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zFCNUW0TfqE.mp4", "duration": 162.0, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a man being electrocuted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a man being electrocuted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 57.0], "base_se_norm": [-0.3827160596847534, -0.29629629850387573], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [-0.3580247163772583, -0.29629629850387573], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000898.pt"}
{"idx": 900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zFCNUW0TfqE.mp4", "duration": 162.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a man in a red shirt and a woman with brown hair are having a serious conversation by the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a man in a red shirt and a woman with brown hair are having a serious conversation by the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 106.0], "base_se_norm": [0.17283952236175537, 0.30864202976226807], "gt_se_raw": [95.0, 105.0], "gt_se_norm": [0.17283952236175537, 0.29629623889923096], "start_token_span": [80, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000900.pt"}
{"idx": 902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/abPLDLV8O4s.mp4", "duration": 102.96, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment where the Santa toy is dancing and playing music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment where the Santa toy is dancing and playing music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 38.0], "base_se_norm": [-0.5726495981216431, -0.2618492841720581], "gt_se_raw": [18.0, 30.0], "gt_se_norm": [-0.6503496170043945, -0.4172494411468506], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000902.pt"}
{"idx": 904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/abPLDLV8O4s.mp4", "duration": 102.96, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat discover and pull out the food from Santa's bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat discover and pull out the food from Santa's bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 52.0], "base_se_norm": [-0.4755244851112366, 0.010100960731506348], "gt_se_raw": [45.0, 53.0], "gt_se_norm": [-0.1258741021156311, 0.0295259952545166], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000904.pt"}
{"idx": 906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/abPLDLV8O4s.mp4", "duration": 102.96, "prompt_text": "Please find the visual event described by the sentence 'When does the cat walk away after eating and playing with all the food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat walk away after eating and playing with all the food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 83.0], "base_se_norm": [0.37917637825012207, 0.6122766733169556], "gt_se_raw": [81.0, 85.0], "gt_se_norm": [0.5734266042709351, 0.6511266231536865], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000906.pt"}
{"idx": 908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MXVz4izTWZY.mp4", "duration": 145.47881214547883, "prompt_text": "Please find the visual event described by the sentence 'Show me the close-up shot of the elevator buttons with braille.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the close-up shot of the elevator buttons with braille.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 38.0], "base_se_norm": [-0.5875687599182129, -0.47758710384368896], "gt_se_raw": [29.0, 38.0], "gt_se_norm": [-0.6013164520263672, -0.47758710384368896], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000908.pt"}
{"idx": 910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MXVz4izTWZY.mp4", "duration": 145.47881214547883, "prompt_text": "Please find the visual event described by the sentence 'Locate the shot of the hospital floor directory.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the shot of the hospital floor directory.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 74.0], "base_se_norm": [-0.09265130758285522, 0.017330288887023926], "gt_se_raw": [66.0, 74.0], "gt_se_norm": [-0.09265130758285522, 0.017330288887023926], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000910.pt"}
{"idx": 912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W6gxb8IunvI.mp4", "duration": 59.62623333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the instructor remove the three blue labels from the map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the instructor remove the three blue labels from the map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 29.0], "base_se_norm": [-0.2285274863243103, -0.02727377414703369], "gt_se_raw": [20.0, 27.0], "gt_se_norm": [-0.3291543126106262, -0.09435832500457764], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000912.pt"}
{"idx": 914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W6gxb8IunvI.mp4", "duration": 59.62623333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the instructor traces the path to the first item, the red curved piece.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the instructor traces the path to the first item, the red curved piece.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [0.1068953275680542, 0.34169137477874756], "gt_se_raw": [31.0, 39.0], "gt_se_norm": [0.039810776710510254, 0.3081490993499756], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000914.pt"}
{"idx": 916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W6gxb8IunvI.mp4", "duration": 59.62623333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the instructor point to the checkout area, represented by the white line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the instructor point to the checkout area, represented by the white line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [0.7441987991333008, 0.8783679008483887], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [0.7777410745620728, 0.8448256254196167], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000916.pt"}
{"idx": 918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4O94zKHSgMU.mp4", "duration": 46.92, "prompt_text": "Please find the visual event described by the sentence 'At what point is the method for visualizing shapes using oil explained?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the method for visualizing shapes using oil explained?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 26.0], "base_se_norm": [-0.10485929250717163, 0.10826945304870605], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [-0.10485929250717163, 0.15089523792266846], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000918.pt"}
{"idx": 920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4O94zKHSgMU.mp4", "duration": 46.92, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip that demonstrates rendering multiple shapes simultaneously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip that demonstrates rendering multiple shapes simultaneously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [0.5345269441604614, 0.8329071998596191], "gt_se_raw": [36.0, 43.0], "gt_se_norm": [0.5345269441604614, 0.8329071998596191], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000920.pt"}
{"idx": 922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mo-U0b64hA.mp4", "duration": 80.0, "prompt_text": "Please find the visual event described by the sentence 'When is the coffee powder being poured into the French press?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the coffee powder being poured into the French press?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.699999988079071, -0.625], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.699999988079071, -0.625], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000922.pt"}
{"idx": 924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mo-U0b64hA.mp4", "duration": 80.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the remaining water added to the French press?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the remaining water added to the French press?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.3999999761581421, -0.25], "gt_se_raw": [25.0, 30.0], "gt_se_norm": [-0.375, -0.25], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000924.pt"}
{"idx": 926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mo-U0b64hA.mp4", "duration": 80.0, "prompt_text": "Please find the visual event described by the sentence 'When is milk poured into the glass of coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is milk poured into the glass of coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [0.2749999761581421, 0.3500000238418579], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [0.2749999761581421, 0.3500000238418579], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000926.pt"}
{"idx": 928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAeAqaA0Llg.mp4", "duration": 152.65249999999997, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a boy in a green and white jacket is called to the front of the class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a boy in a green and white jacket is called to the front of the class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.7117636203765869, -0.6593570709228516], "gt_se_raw": [75.0, 81.0], "gt_se_norm": [-0.017376065254211426, 0.06123387813568115], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000928.pt"}
{"idx": 930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAeAqaA0Llg.mp4", "duration": 152.65249999999997, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the suit ask the students to take their seats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the suit ask the students to take their seats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 63.0], "base_se_norm": [-0.3318157196044922, -0.1745958924293518], "gt_se_raw": [58.0, 63.0], "gt_se_norm": [-0.24010413885116577, -0.1745958924293518], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000930.pt"}
{"idx": 932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qOZNzVcq2IY.mp4", "duration": 174.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a visually impaired man getting off the bus with a cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a visually impaired man getting off the bus with a cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.9770159125328064, -0.9540318250656128], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9770159125328064, -0.9540318250656128], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000932.pt"}
{"idx": 934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qOZNzVcq2IY.mp4", "duration": 174.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the Siri interface shown as an example of a voice-activated system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the Siri interface shown as an example of a voice-activated system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 62.0], "base_se_norm": [-0.40241336822509766, -0.28749287128448486], "gt_se_raw": [52.0, 62.0], "gt_se_norm": [-0.40241336822509766, -0.28749287128448486], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000934.pt"}
{"idx": 936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qOZNzVcq2IY.mp4", "duration": 174.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a person in a striped shirt hails and gets on a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a person in a striped shirt hails and gets on a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 11.0], "base_se_norm": [-0.9425397515296936, -0.873587429523468], "gt_se_raw": [113.0, 120.0], "gt_se_norm": [0.29860174655914307, 0.37904608249664307], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000936.pt"}
{"idx": 938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ke6s5UH9n2w.mp4", "duration": 162.42893333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the player types 'i have a song for you my comrade'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the player types 'i have a song for you my comrade'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.5813553929328918, -0.5321030616760254], "gt_se_raw": [38.0, 39.0], "gt_se_norm": [-0.5321030616760254, -0.51978999376297], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000938.pt"}
{"idx": 940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ke6s5UH9n2w.mp4", "duration": 162.42893333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player wearing a blue helmet and purple armor bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player wearing a blue helmet and purple armor bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [-0.7044861316680908, -0.67985999584198], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.04661154747009277, 0.07123768329620361], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000940.pt"}
{"idx": 942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOKLZsyN-aw.mp4", "duration": 14.6, "prompt_text": "Please find the visual event described by the sentence 'When is the dog standing on a rock overlooking the city?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog standing on a rock overlooking the city?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.9178081750869751], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, 0.9178081750869751], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000942.pt"}
{"idx": 944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOKLZsyN-aw.mp4", "duration": 14.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the interval where the dog is facing forward and has its mouth open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the interval where the dog is facing forward and has its mouth open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.31506848335266113, 0.09589040279388428], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.31506848335266113, 0.23287665843963623], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000944.pt"}
{"idx": 946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOKLZsyN-aw.mp4", "duration": 14.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the dog is looking towards the left side of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the dog is looking towards the left side of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.45205479860305786], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.45205479860305786], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000946.pt"}
{"idx": 948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPQwSEzXBXU.mp4", "duration": 104.4, "prompt_text": "Please find the visual event described by the sentence 'Show the time when people are grading papers at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when people are grading papers at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [0.034482717514038086, 0.14942526817321777], "gt_se_raw": [54.0, 60.0], "gt_se_norm": [0.034482717514038086, 0.14942526817321777], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000948.pt"}
{"idx": 950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPQwSEzXBXU.mp4", "duration": 104.4, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man in a blue shirt assists a boy in an orange shirt with a braille typewriter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man in a blue shirt assists a boy in an orange shirt with a braille typewriter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 65.0], "base_se_norm": [0.14942526817321777, 0.24521076679229736], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [0.14942526817321777, 0.24521076679229736], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000950.pt"}
{"idx": 952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0klQnTscgRs.mp4", "duration": 117.05026666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a black suit introducing a pink device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a black suit introducing a pink device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.7778732180595398, -0.6924397945404053], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.7949599027633667, -0.709526538848877], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000952.pt"}
{"idx": 954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0klQnTscgRs.mp4", "duration": 117.05026666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the pink attachment flash a red light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the pink attachment flash a red light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.8462198972702026, -0.7949599027633667], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.5044864416122437, -0.45322638750076294], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000954.pt"}
{"idx": 956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0klQnTscgRs.mp4", "duration": 117.05026666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with a person doing yoga in the blurred background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with a person doing yoga in the blurred background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [0.2985873222351074, 0.38402068614959717], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [0.2985873222351074, 0.38402068614959717], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000956.pt"}
{"idx": 958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Zh2-rVsXWUU.mp4", "duration": 107.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the two superheroes are arguing in the middle of the fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the two superheroes are arguing in the middle of the fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 39.0], "base_se_norm": [-0.7012696266174316, -0.27184468507766724], "gt_se_raw": [17.0, 32.0], "gt_se_norm": [-0.6825989484786987, -0.4025392532348633], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000958.pt"}
{"idx": 960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Zh2-rVsXWUU.mp4", "duration": 107.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'A man in a green ski mask gets a drink of water while a police officer points a gun at him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a green ski mask gets a drink of water while a police officer points a gun at him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 86.0], "base_se_norm": [0.5309932231903076, 0.6056758165359497], "gt_se_raw": [71.0, 91.0], "gt_se_norm": [0.32561612129211426, 0.6990290880203247], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000960.pt"}
{"idx": 962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cT_Wuzag6VU.mp4", "duration": 129.66286666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man walking towards the edge of a cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man walking towards the edge of a cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 23.0], "base_se_norm": [-0.7994799613952637, -0.6452338695526123], "gt_se_raw": [1.0, 16.0], "gt_se_norm": [-0.9845753908157349, -0.7532061338424683], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000962.pt"}
{"idx": 964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cT_Wuzag6VU.mp4", "duration": 129.66286666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man use his hands to brake while skateboarding around a sharp turn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man use his hands to brake while skateboarding around a sharp turn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 89.0], "base_se_norm": [0.3110923767089844, 0.3727908134460449], "gt_se_raw": [84.0, 88.0], "gt_se_norm": [0.29566776752471924, 0.3573662042617798], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000964.pt"}
{"idx": 966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cT_Wuzag6VU.mp4", "duration": 129.66286666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man skate into a town and stop in front of a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man skate into a town and stop in front of a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 128.0], "base_se_norm": [0.820104718208313, 0.9743508100509644], "gt_se_raw": [120.0, 126.0], "gt_se_norm": [0.8509539365768433, 0.9435015916824341], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000966.pt"}
{"idx": 968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AMHWyS5GyeY.mp4", "duration": 101.68, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man tries to coax the cat into the carrier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man tries to coax the cat into the carrier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 32.0], "base_se_norm": [-0.8426436185836792, -0.37057435512542725], "gt_se_raw": [15.0, 26.0], "gt_se_norm": [-0.7049567699432373, -0.48859167098999023], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000968.pt"}
{"idx": 970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AMHWyS5GyeY.mp4", "duration": 101.68, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when Simon's cat sniffs the kitten.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when Simon's cat sniffs the kitten.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.16050350666046143, 0.2391817569732666], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [0.1998426914215088, 0.2588512897491455], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000970.pt"}
{"idx": 972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzvK1lJwVUM.mp4", "duration": 133.2164111111111, "prompt_text": "Please find the visual event described by the sentence 'When are the eggs being added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the eggs being added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.7898156642913818, -0.74477618932724], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.7898156642913818, -0.74477618932724], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000972.pt"}
{"idx": 974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzvK1lJwVUM.mp4", "duration": 133.2164111111111, "prompt_text": "Please find the visual event described by the sentence 'When are the M&M-like chocolate candies added to the cookie dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the M&M-like chocolate candies added to the cookie dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.21931540966033936, -0.1592627763748169], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [-0.21931540966033936, -0.1742759346961975], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000974.pt"}
{"idx": 976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzvK1lJwVUM.mp4", "duration": 133.2164111111111, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the baked cookie is being sliced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the baked cookie is being sliced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 102.0], "base_se_norm": [0.38121116161346436, 0.5313428640365601], "gt_se_raw": [95.0, 100.0], "gt_se_norm": [0.426250696182251, 0.5013164281845093], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000976.pt"}
{"idx": 978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ftde81QNXDY.mp4", "duration": 6.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the doll's recorded voice can be heard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the doll's recorded voice can be heard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, 0.7733989953994751], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.11330050230026245, 0.7733989953994751], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000978.pt"}
{"idx": 980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ftde81QNXDY.mp4", "duration": 6.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the exact moment the doll says the phrase 'I'm just a girl'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the exact moment the doll says the phrase 'I'm just a girl'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.4088670015335083, 0.1822659969329834], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [0.1822659969329834, 0.7733989953994751], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000980.pt"}
{"idx": 982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WwPXYlVwnk.mp4", "duration": 76.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a purple hijab and a woman in purple armor standing together outdoors near solar panels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a purple hijab and a woman in purple armor standing together outdoors near solar panels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.8947368264198303, -0.8421052694320679], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8947368264198303, -0.8421052694320679], "start_token_span": [78, 81], "end_token_span": [83, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000982.pt"}
{"idx": 984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WwPXYlVwnk.mp4", "duration": 76.0, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the character in purple runs through a series of red laser beams.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the character in purple runs through a series of red laser beams.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [-0.15789473056793213, -0.026315808296203613], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.13157892227172852, -0.05263155698776245], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000984.pt"}
{"idx": 986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WwPXYlVwnk.mp4", "duration": 76.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the character jumps onto a pole in a city setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the character jumps onto a pole in a city setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 68.0], "base_se_norm": [0.736842155456543, 0.7894736528396606], "gt_se_raw": [67.0, 68.0], "gt_se_norm": [0.763157844543457, 0.7894736528396606], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000986.pt"}
{"idx": 988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJry7As0rJU.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman in a black shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman in a black shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [-0.5677399635314941, -0.48368942737579346], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.5557327270507812, -0.48368942737579346], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000988.pt"}
{"idx": 990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJry7As0rJU.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman wearing a blue sweater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman wearing a blue sweater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 68.0], "base_se_norm": [-0.25555217266082764, -0.18350881338119507], "gt_se_raw": [62.0, 69.0], "gt_se_norm": [-0.25555217266082764, -0.17150157690048218], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000990.pt"}
{"idx": 992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUusX1Js6R0.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When does Bert express his frustration about not catching any fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Bert express his frustration about not catching any fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 61.0], "base_se_norm": [-0.5677399635314941, -0.2675594091415405], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.8679205179214478, -0.7838699817657471], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000992.pt"}
{"idx": 994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUusX1Js6R0.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point do multiple fish jump into the boat at once?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 165.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do multiple fish jump into the boat at once?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 165.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 165.0], "base_se_norm": [0.9091484546661377, 0.9811917543411255], "gt_se_raw": [91.0, 96.0], "gt_se_norm": [0.09265732765197754, 0.15269339084625244], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000994.pt"}
{"idx": 996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUusX1Js6R0.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the shark appear and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 166.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the shark appear and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 166.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 166.0], "base_se_norm": [0.825097918510437, 0.9931989908218384], "gt_se_raw": [150.0, 155.0], "gt_se_norm": [0.8010834455490112, 0.8611196279525757], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000996.pt"}
{"idx": 998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0LHxvxdRnYc.mp4", "duration": 138.30483333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot where the woman's body is covered in blue electric energy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot where the woman's body is covered in blue electric energy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [-0.36372435092926025, -0.32034194469451904], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.36372435092926025, -0.32034194469451904], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000998.pt"}
{"idx": 1000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0LHxvxdRnYc.mp4", "duration": 138.30483333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where the woman with glowing eyes is floating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where the woman with glowing eyes is floating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [-0.36372435092926025, -0.32034194469451904], "gt_se_raw": [100.0, 101.0], "gt_se_norm": [0.4460810422897339, 0.46054184436798096], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001000.pt"}
{"idx": 1002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nr1IhBX8m1w.mp4", "duration": 28.5, "prompt_text": "Please find the visual event described by the sentence 'When is a person seen walking on the platform?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person seen walking on the platform?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.4385964870452881], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5087718963623047], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001002.pt"}
{"idx": 1004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nr1IhBX8m1w.mp4", "duration": 28.5, "prompt_text": "Please find the visual event described by the sentence 'What is the duration of the train passing directly in front of the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the duration of the train passing directly in front of the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 19.0], "base_se_norm": [-0.08771932125091553, 0.3333333730697632], "gt_se_raw": [13.0, 20.0], "gt_se_norm": [-0.08771932125091553, 0.4035087823867798], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001004.pt"}
{"idx": 1006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nr1IhBX8m1w.mp4", "duration": 28.5, "prompt_text": "Please find the visual event described by the sentence 'Show the train after it has passed the station and is moving away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the train after it has passed the station and is moving away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 28.0], "base_se_norm": [0.3333333730697632, 0.9649122953414917], "gt_se_raw": [20.0, 28.0], "gt_se_norm": [0.4035087823867798, 0.9649122953414917], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001006.pt"}
{"idx": 1008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_B2UZ_ZoxU.mp4", "duration": 185.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do the trophies get knocked off the shelves by the ink blobs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the trophies get knocked off the shelves by the ink blobs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.5267757177352905, -0.48375529050827026], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [-0.505265474319458, -0.48375529050827026], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001008.pt"}
{"idx": 1010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_B2UZ_ZoxU.mp4", "duration": 185.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the small ink blobs merge and transform into a large monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the small ink blobs merge and transform into a large monster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 108.0], "base_se_norm": [0.10777509212493896, 0.16155052185058594], "gt_se_raw": [103.0, 107.0], "gt_se_norm": [0.10777509212493896, 0.15079545974731445], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001010.pt"}
{"idx": 1012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J95dkdlWQbw.mp4", "duration": 94.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When do we see the man in the red outfit start dancing and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see the man in the red outfit start dancing and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.7447713613510132, -0.6596951484680176], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.7447713613510132, -0.6809642314910889], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001012.pt"}
{"idx": 1014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J95dkdlWQbw.mp4", "duration": 94.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a suit shown giving a speech on television?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a suit shown giving a speech on television?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [-0.19177597761154175, -0.08543068170547485], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [-0.19177597761154175, -0.08543068170547485], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001014.pt"}
{"idx": 1016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J95dkdlWQbw.mp4", "duration": 94.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment on the television where a man is eating a piece of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment on the television where a man is eating a piece of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [-0.06416165828704834, 0.08472168445587158], "gt_se_raw": [69.0, 73.0], "gt_se_norm": [0.4675647020339966, 0.5526409149169922], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001016.pt"}
{"idx": 1018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hywRdDVR76A.mp4", "duration": 135.52, "prompt_text": "Please find the visual event described by the sentence 'At what point is the double helix structure of DNA demonstrated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the double helix structure of DNA demonstrated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 76.0], "base_se_norm": [-0.07024794816970825, 0.12160563468933105], "gt_se_raw": [70.0, 75.0], "gt_se_norm": [0.0330578088760376, 0.10684764385223389], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001018.pt"}
{"idx": 1020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hywRdDVR76A.mp4", "duration": 135.52, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the four bases of DNA (A, T, G, C) are introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the four bases of DNA (A, T, G, C) are introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 84.0], "base_se_norm": [0.1511216163635254, 0.2396693229675293], "gt_se_raw": [76.0, 84.0], "gt_se_norm": [0.12160563468933105, 0.2396693229675293], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001020.pt"}
{"idx": 1022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DIBw9dSVKdU.mp4", "duration": 126.25112499999999, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man connect the computer to the internet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man connect the computer to the internet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.6990125775337219, -0.6514882445335388], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.6831711530685425, -0.6514882445335388], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001022.pt"}
{"idx": 1024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DIBw9dSVKdU.mp4", "duration": 126.25112499999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when Ralph is walking through the crowd and getting distracted by pop-up ads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when Ralph is walking through the crowd and getting distracted by pop-up ads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 60.0], "base_se_norm": [-0.16040349006652832, -0.049513399600982666], "gt_se_raw": [53.0, 60.0], "gt_se_norm": [-0.16040349006652832, -0.049513399600982666], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001024.pt"}
{"idx": 1026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DIBw9dSVKdU.mp4", "duration": 126.25112499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the bunny explode from eating too many pancakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bunny explode from eating too many pancakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 106.0], "base_se_norm": [0.6316686868667603, 0.6791930198669434], "gt_se_raw": [112.0, 116.0], "gt_se_norm": [0.7742416858673096, 0.8376073837280273], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001026.pt"}
{"idx": 1028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DqilSuFK3B8.mp4", "duration": 438.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the woman in the plaid top sitting on the bed and speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the woman in the plaid top sitting on the bed and speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 45.0], "base_se_norm": [-0.9088007211685181, -0.794801652431488], "gt_se_raw": [21.0, 46.0], "gt_se_norm": [-0.9042407870292664, -0.7902416586875916], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001028.pt"}
{"idx": 1030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DqilSuFK3B8.mp4", "duration": 438.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the segment showing the bloopers or failed dance moves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 330.0 - 353.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the segment showing the bloopers or failed dance moves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 330.0 - 353.0 seconds.<|im_end|>\n", "base_se_raw": [330.0, 353.0], "base_se_norm": [0.5047879219055176, 0.6096670627593994], "gt_se_raw": [335.0, 351.0], "gt_se_norm": [0.5275877714157104, 0.600547194480896], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001030.pt"}
{"idx": 1032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5L4DQfVIcdg.mp4", "duration": 314.875, "prompt_text": "Please find the visual event described by the sentence 'When does the boy imagine his toy rocket is flying in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy imagine his toy rocket is flying in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 12.0], "base_se_norm": [-0.9872965216636658, -0.9237792491912842], "gt_se_raw": [2.0, 12.0], "gt_se_norm": [-0.9872965216636658, -0.9237792491912842], "start_token_span": [74, 77], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001032.pt"}
{"idx": 1034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5L4DQfVIcdg.mp4", "duration": 314.875, "prompt_text": "Please find the visual event described by the sentence 'Where in the video does the boy build and set up his lemonade stand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video does the boy build and set up his lemonade stand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 150.0], "base_se_norm": [-0.1488686203956604, -0.0472409725189209], "gt_se_raw": [133.0, 149.0], "gt_se_norm": [-0.15522032976150513, -0.053592681884765625], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001034.pt"}
{"idx": 1036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5L4DQfVIcdg.mp4", "duration": 314.875, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the old man pours a bag of coins into the coin slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 207.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the old man pours a bag of coins into the coin slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 207.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [207.0, 222.0], "base_se_norm": [0.3148074150085449, 0.4100834131240845], "gt_se_raw": [211.0, 218.0], "gt_se_norm": [0.3402143716812134, 0.384676456451416], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001036.pt"}
{"idx": 1038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T52u9jDtIM4.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where students are seated in the Argo Central computer lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where students are seated in the Argo Central computer lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.703417181968689, -0.6629740595817566], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.7168982028961182, -0.6764551401138306], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001038.pt"}
{"idx": 1040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T52u9jDtIM4.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a slide with the title 'Argo 30 Guarantee Requirements' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a slide with the title 'Argo 30 Guarantee Requirements' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 71.0], "base_se_norm": [-0.16417568922042847, -0.04284638166427612], "gt_se_raw": [62.0, 71.0], "gt_se_norm": [-0.16417568922042847, -0.04284638166427612], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001040.pt"}
{"idx": 1042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2hm0_sNiN2M.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone rubbing their soapy hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone rubbing their soapy hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 24.0], "base_se_norm": [-0.3999999761581421, -0.19999998807907104], "gt_se_raw": [18.0, 24.0], "gt_se_norm": [-0.3999999761581421, -0.19999998807907104], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001042.pt"}
{"idx": 1044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2hm0_sNiN2M.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'At what point are the soapy hands being rinsed under running water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the soapy hands being rinsed under running water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [0.2666666507720947, 0.36666667461395264], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [0.2666666507720947, 0.36666667461395264], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001044.pt"}
{"idx": 1046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2hm0_sNiN2M.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the paper towel is used to turn off the faucet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the paper towel is used to turn off the faucet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [0.6333333253860474, 0.7000000476837158], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [0.6333333253860474, 0.7000000476837158], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001046.pt"}
{"idx": 1048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I8ZvdblLcnk.mp4", "duration": 63.998999999999995, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl tries to put the sock on by herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl tries to put the sock on by herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 35.0], "base_se_norm": [-0.624994158744812, 0.09376704692840576], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.5937436819076538, -0.4687417149543762], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001048.pt"}
{"idx": 1050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I8ZvdblLcnk.mp4", "duration": 63.998999999999995, "prompt_text": "Please find the visual event described by the sentence 'When does the girl successfully pull the sock over her heel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl successfully pull the sock over her heel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [0.7812777757644653, 0.9687807559967041], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [0.7500272989273071, 0.8437788486480713], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001050.pt"}
{"idx": 1052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WoKaJfUHJmg.mp4", "duration": 166.09926666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman wave to the camera in front of the Opera House?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman wave to the camera in front of the Opera House?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 11.0], "base_se_norm": [-0.9638770222663879, -0.8675490617752075], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.9518360495567322, -0.9157130718231201], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001052.pt"}
{"idx": 1054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WoKaJfUHJmg.mp4", "duration": 166.09926666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a puppet show with a red crab and a green turtle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a puppet show with a red crab and a green turtle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 82.0], "base_se_norm": [-0.04876166582107544, -0.012638688087463379], "gt_se_raw": [79.0, 82.0], "gt_se_norm": [-0.04876166582107544, -0.012638688087463379], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001054.pt"}
{"idx": 1056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WoKaJfUHJmg.mp4", "duration": 166.09926666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman stand in front of the lift entrance, pointing upwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman stand in front of the lift entrance, pointing upwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 123.0], "base_se_norm": [0.45695996284484863, 0.4810420274734497], "gt_se_raw": [122.0, 124.0], "gt_se_norm": [0.46900105476379395, 0.49308300018310547], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001056.pt"}
{"idx": 1058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1lETr5TeT7E.mp4", "duration": 106.53976666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl steps into the rocky area of the fountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl steps into the rocky area of the fountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.7559596300125122, -0.6620979905128479], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.2303343415260315, -0.1364726424217224], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001058.pt"}
{"idx": 1060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1lETr5TeT7E.mp4", "duration": 106.53976666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where the girl stands up, looks up, and seems to feel the wind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where the girl stands up, looks up, and seems to feel the wind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.887365996837616, -0.7935043573379517], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [0.27651870250701904, 0.3703804016113281], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001060.pt"}
{"idx": 1062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TU6VrllUQjw.mp4", "duration": 29.462766666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man wearing red suspenders and a white shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man wearing red suspenders and a white shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.3211771249771118], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, -0.3211771249771118], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001062.pt"}
{"idx": 1064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TU6VrllUQjw.mp4", "duration": 29.462766666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the blue banner with phone numbers is visible at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the blue banner with phone numbers is visible at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 29.0], "base_se_norm": [-1.0, 0.9685863256454468], "gt_se_raw": [0.0, 29.0], "gt_se_norm": [-1.0, 0.9685863256454468], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001064.pt"}
{"idx": 1066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TU6VrllUQjw.mp4", "duration": 29.462766666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a wide shot shows four men sitting at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a wide shot shows four men sitting at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 29.0], "base_se_norm": [0.900704026222229, 0.9685863256454468], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.8328217267990112, 0.9685863256454468], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001066.pt"}
{"idx": 1068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/olit-B5Yldc.mp4", "duration": 108.17391304347827, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a mother gets angry with her crying baby at a dinner table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a mother gets angry with her crying baby at a dinner table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.5008038282394409, -0.3713826537132263], "gt_se_raw": [27.0, 33.0], "gt_se_norm": [-0.5008038282394409, -0.38987135887145996], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001068.pt"}
{"idx": 1070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/olit-B5Yldc.mp4", "duration": 108.17391304347827, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where an old man shouts at people from inside an elevator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where an old man shouts at people from inside an elevator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 73.0], "base_se_norm": [0.2942122220993042, 0.34967851638793945], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [0.3127009868621826, 0.34967851638793945], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001070.pt"}
{"idx": 1072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZbCE6wzNbE.mp4", "duration": 105.8, "prompt_text": "Please find the visual event described by the sentence 'At what point is the Bookshare Web Reader interface with the math problem displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the Bookshare Web Reader interface with the math problem displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 40.0], "base_se_norm": [-0.7164461612701416, -0.24385637044906616], "gt_se_raw": [15.0, 40.0], "gt_se_norm": [-0.7164461612701416, -0.24385637044906616], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001072.pt"}
{"idx": 1074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZbCE6wzNbE.mp4", "duration": 105.8, "prompt_text": "Please find the visual event described by the sentence 'When are the logos of Microsoft, the American Mathematical Society, and ChromeVox shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the logos of Microsoft, the American Mathematical Society, and ChromeVox shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 50.0], "base_se_norm": [-0.22495275735855103, -0.05482041835784912], "gt_se_raw": [42.0, 50.0], "gt_se_norm": [-0.2060491442680359, -0.05482041835784912], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001074.pt"}
{"idx": 1076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZbCE6wzNbE.mp4", "duration": 105.8, "prompt_text": "Please find the visual event described by the sentence 'When does the demonstration of the MathML Cloud's web interface generating the equation happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the demonstration of the MathML Cloud's web interface generating the equation happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 74.0], "base_se_norm": [0.15311908721923828, 0.3988656997680664], "gt_se_raw": [61.0, 70.0], "gt_se_norm": [0.15311908721923828, 0.3232513666152954], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001076.pt"}
{"idx": 1078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G1LP4CwL0KE.mp4", "duration": 46.6, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher fan out a set of number cards for the boy to choose from?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher fan out a set of number cards for the boy to choose from?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 33.0], "base_se_norm": [-0.09871244430541992, 0.416309118270874], "gt_se_raw": [22.0, 29.0], "gt_se_norm": [-0.055793941020965576, 0.2446352243423462], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001078.pt"}
{"idx": 1080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G1LP4CwL0KE.mp4", "duration": 46.6, "prompt_text": "Please find the visual event described by the sentence 'At what time does the teacher lean in close to the boy's ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the teacher lean in close to the boy's ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [0.8454936742782593, 0.974249005317688], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [0.8454936742782593, 0.974249005317688], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001080.pt"}
{"idx": 1082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TubxNbCQ4Fk.mp4", "duration": 171.90506666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two men riding a jet ski on the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men riding a jet ski on the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.90692538022995, -0.8254850506782532], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.90692538022995, -0.8603880405426025], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001082.pt"}
{"idx": 1084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TubxNbCQ4Fk.mp4", "duration": 171.90506666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a man surfing inside a wave tube?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a man surfing inside a wave tube?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 93.0], "base_se_norm": [0.0005522966384887695, 0.08199262619018555], "gt_se_raw": [86.0, 92.0], "gt_se_norm": [0.0005522966384887695, 0.0703582763671875], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001084.pt"}
{"idx": 1086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TubxNbCQ4Fk.mp4", "duration": 171.90506666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two men walking on the railway track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men walking on the railway track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [-0.627701461315155, -0.5695298314094543], "gt_se_raw": [147.0, 149.0], "gt_se_norm": [0.7102463245391846, 0.7335150241851807], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001086.pt"}
{"idx": 1088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aSq1cez_flQ.mp4", "duration": 133.71810027119116, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman sings her first solo verse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman sings her first solo verse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 43.0], "base_se_norm": [-0.6859064102172852, -0.35685592889785767], "gt_se_raw": [24.0, 45.0], "gt_se_norm": [-0.6410358548164368, -0.3269422650337219], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001088.pt"}
{"idx": 1090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aSq1cez_flQ.mp4", "duration": 133.71810027119116, "prompt_text": "Please find the visual event described by the sentence 'When are the man and woman singing in unison?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and woman singing in unison?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 63.0], "base_se_norm": [-0.2521580457687378, -0.05771917104721069], "gt_se_raw": [93.0, 110.0], "gt_se_norm": [0.3909860849380493, 0.6452523469924927], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001090.pt"}
{"idx": 1092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h6oooSMEeQw.mp4", "duration": 106.87343333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene with the Colosseum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene with the Colosseum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 88.0], "base_se_norm": [0.5345253944396973, 0.6468076705932617], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [0.553239107131958, 0.6468076705932617], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001092.pt"}
{"idx": 1094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h6oooSMEeQw.mp4", "duration": 106.87343333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the sheet music is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the sheet music is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 97.0], "base_se_norm": [0.6468076705932617, 0.815231204032898], "gt_se_raw": [88.0, 98.0], "gt_se_norm": [0.6468076705932617, 0.8339449167251587], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001094.pt"}
{"idx": 1096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Co4dLH29PvM.mp4", "duration": 89.58949999999999, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman in a lab coat is performing an experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman in a lab coat is performing an experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.7767595648765564, -0.6651393175125122], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.7767595648765564, -0.6874634027481079], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001096.pt"}
{"idx": 1098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Co4dLH29PvM.mp4", "duration": 89.58949999999999, "prompt_text": "Please find the visual event described by the sentence 'Find the time when a man and a woman are having a conversation while seated in chairs near a large window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when a man and a woman are having a conversation while seated in chairs near a large window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.843731701374054, -0.7767595648765564], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.843731701374054, -0.7767595648765564], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001098.pt"}
{"idx": 1100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6wFr-o_M_Uo.mp4", "duration": 122.8, "prompt_text": "Please find the visual event described by the sentence 'When does the woman state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9674267172813416, -0.9185667634010315], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9348534345626831, -0.9022801518440247], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001100.pt"}
{"idx": 1102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6wFr-o_M_Uo.mp4", "duration": 122.8, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker talk about Android phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker talk about Android phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.6742671132087708, -0.5114006996154785], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.4951140284538269, -0.44625407457351685], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001102.pt"}
{"idx": 1104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6wFr-o_M_Uo.mp4", "duration": 122.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman provides the email address.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman provides the email address.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 94.0], "base_se_norm": [0.465798020362854, 0.5309445858001709], "gt_se_raw": [109.0, 114.0], "gt_se_norm": [0.7752442359924316, 0.8566775321960449], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001104.pt"}
{"idx": 1106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DvaPRlZtfyc.mp4", "duration": 106.9068, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when an ambulance appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when an ambulance appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.7755054235458374, -0.7380896210670471], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.7755054235458374, -0.7193817496299744], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001106.pt"}
{"idx": 1108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DvaPRlZtfyc.mp4", "duration": 106.9068, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where Mario in a wheelchair goes down an elevator at the flagpole.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where Mario in a wheelchair goes down an elevator at the flagpole.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 79.0], "base_se_norm": [0.3469676971435547, 0.47792279720306396], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [0.3469676971435547, 0.3843834400177002], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001108.pt"}
{"idx": 1110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKTQW5i9_p8.mp4", "duration": 125.44, "prompt_text": "Please find the visual event described by the sentence 'When does the girl introduce herself as Greta Thunberg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl introduce herself as Greta Thunberg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9840561151504517, -0.9202806353569031], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9840561151504517, -0.9681122303009033], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001110.pt"}
{"idx": 1112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKTQW5i9_p8.mp4", "duration": 125.44, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the crowd gives a standing ovation after the speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the crowd gives a standing ovation after the speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 51.0], "base_se_norm": [-0.23469388484954834, -0.1868622899055481], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.23469388484954834, -0.1868622899055481], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001112.pt"}
{"idx": 1114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKTQW5i9_p8.mp4", "duration": 125.44, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where she says 'How dare you'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where she says 'How dare you'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 107.0], "base_se_norm": [0.6422193050384521, 0.7059948444366455], "gt_se_raw": [103.0, 105.0], "gt_se_norm": [0.6422193050384521, 0.6741070747375488], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001114.pt"}
{"idx": 1116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X-AldaPHQdE.mp4", "duration": 107.07363333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy demonstrates that the egg cannot fit into the bottle by itself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy demonstrates that the egg cannot fit into the bottle by itself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 54.0], "base_se_norm": [-0.06606322526931763, 0.0086517333984375], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.04738450050354004, -0.010027050971984863], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001116.pt"}
{"idx": 1118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X-AldaPHQdE.mp4", "duration": 107.07363333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the egg is sucked into the bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the egg is sucked into the bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 75.0], "base_se_norm": [0.17676031589508057, 0.40090513229370117], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [0.17676031589508057, 0.2327965497970581], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001118.pt"}
{"idx": 1120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRjCaSrmm2A.mp4", "duration": 53.13641666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the friends discussing the rules of the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the friends discussing the rules of the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9623610377311707, -0.7741661667823792], "gt_se_raw": [7.0, 26.0], "gt_se_norm": [-0.7365272045135498, -0.021386802196502686], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001120.pt"}
{"idx": 1122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRjCaSrmm2A.mp4", "duration": 53.13641666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a hammer seen lodged in a tree stump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a hammer seen lodged in a tree stump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [0.01625216007232666, 0.09153008460998535], "gt_se_raw": [27.0, 28.0], "gt_se_norm": [0.01625216007232666, 0.05389118194580078], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001122.pt"}
{"idx": 1124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRjCaSrmm2A.mp4", "duration": 53.13641666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the person in the red shirt and mask scream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in the red shirt and mask scream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [0.46791982650756836, 0.543197751045227], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [0.43028080463409424, 0.543197751045227], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001124.pt"}
{"idx": 1126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L8qOb8mdTTU.mp4", "duration": 146.56, "prompt_text": "Please find the visual event described by the sentence 'Show me when the cat is kneading the laundry before taking a nap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the cat is kneading the laundry before taking a nap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 24.0], "base_se_norm": [-0.8225982189178467, -0.6724890470504761], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.5906113386154175, -0.5360261797904968], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001126.pt"}
{"idx": 1128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L8qOb8mdTTU.mp4", "duration": 146.56, "prompt_text": "Please find the visual event described by the sentence 'At what point does the violently shaking washing machine cause the cat and laundry to fall off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the violently shaking washing machine cause the cat and laundry to fall off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 116.0], "base_se_norm": [0.501091718673706, 0.5829694271087646], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [0.501091718673706, 0.5283843278884888], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001128.pt"}
{"idx": 1130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HodFzcJHIYI.mp4", "duration": 188.48, "prompt_text": "Please find the visual event described by the sentence 'When are the football players celebrating a touchdown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the football players celebrating a touchdown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.8514431118965149, -0.8302206993103027], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.8514431118965149, -0.8302206993103027], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001130.pt"}
{"idx": 1132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HodFzcJHIYI.mp4", "duration": 188.48, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a man is being pushed in a wheelchair through a hospital corridor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a man is being pushed in a wheelchair through a hospital corridor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.5012733340263367, -0.44821733236312866], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.4906620979309082, -0.4694397449493408], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001132.pt"}
{"idx": 1134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HodFzcJHIYI.mp4", "duration": 188.48, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a coach tells a player he's now a pit bull, not a gazelle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a coach tells a player he's now a pit bull, not a gazelle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 135.0], "base_se_norm": [0.4006791114807129, 0.4325127601623535], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.018675804138183594, 0.03989815711975098], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001134.pt"}
{"idx": 1136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-Kb1VHx2pKk.mp4", "duration": 171.03753333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the problem 468 minus 299 is being calculated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the problem 468 minus 299 is being calculated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 120.0], "base_se_norm": [0.05240058898925781, 0.40320074558258057], "gt_se_raw": [11.0, 40.0], "gt_se_norm": [-0.8713732957839966, -0.53226637840271], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001136.pt"}
{"idx": 1138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-Kb1VHx2pKk.mp4", "duration": 171.03753333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the calculation for 1346 minus 369 is performed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the calculation for 1346 minus 369 is performed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 120.0], "base_se_norm": [0.05240058898925781, 0.40320074558258057], "gt_se_raw": [96.0, 122.0], "gt_se_norm": [0.12256062030792236, 0.42658746242523193], "start_token_span": [76, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001138.pt"}
{"idx": 1140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uceySVBjKNw.mp4", "duration": 89.52276666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman helping the boy with his seatbelt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman helping the boy with his seatbelt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 21.0], "base_se_norm": [-1.0, -0.530845582485199], "gt_se_raw": [3.0, 22.0], "gt_se_norm": [-0.9329779148101807, -0.5085048675537109], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001140.pt"}
{"idx": 1142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uceySVBjKNw.mp4", "duration": 89.52276666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the wagon is pulled past the numbers on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the wagon is pulled past the numbers on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [-0.2850980758666992, -0.17339462041854858], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.2627573609352112, -0.15105390548706055], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001142.pt"}
{"idx": 1144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uceySVBjKNw.mp4", "duration": 89.52276666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the woman and boy passing the cross-shaped art on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman and boy passing the cross-shaped art on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [0.07235288619995117, 0.18405640125274658], "gt_se_raw": [61.0, 70.0], "gt_se_norm": [0.3627818822860718, 0.5638480186462402], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001144.pt"}
{"idx": 1146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h4QzRfvkJZ4.mp4", "duration": 157.69920833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman in the pink shirt enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman in the pink shirt enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.5687993764877319, -0.518069863319397], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.5814816951751709, -0.518069863319397], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001146.pt"}
{"idx": 1148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h4QzRfvkJZ4.mp4", "duration": 157.69920833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the screen display the text 'THE MORNING AFTER'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the screen display the text 'THE MORNING AFTER'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 112.0], "base_se_norm": [0.3950608968734741, 0.4204256534576416], "gt_se_raw": [110.0, 111.0], "gt_se_norm": [0.3950608968734741, 0.40774333477020264], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001148.pt"}
{"idx": 1150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dh1V8GyyNYE.mp4", "duration": 142.85104166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man in the black hat falls from the broken ladder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man in the black hat falls from the broken ladder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.7619898319244385, -0.7059874534606934], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.7479892373085022, -0.7059874534606934], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001150.pt"}
{"idx": 1152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dh1V8GyyNYE.mp4", "duration": 142.85104166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two men running towards the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two men running towards the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.3419719338417053, -0.27196890115737915], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.32797127962112427, -0.2999700903892517], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001152.pt"}
{"idx": 1154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dh1V8GyyNYE.mp4", "duration": 142.85104166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the large black pipe hits both men, causing them to fall through the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the large black pipe hits both men, causing them to fall through the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 94.0], "base_se_norm": [0.23205256462097168, 0.31605613231658936], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [0.2880549430847168, 0.3300567865371704], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001154.pt"}
{"idx": 1156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzrlM6uX9oU.mp4", "duration": 93.72706039372706, "prompt_text": "Please find the visual event described by the sentence 'When is the logo 'S.W.I.M. T.E.A.M.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the logo 'S.W.I.M. T.E.A.M.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 41.0], "base_se_norm": [-0.2744891047477722, -0.12511926889419556], "gt_se_raw": [33.0, 41.0], "gt_se_norm": [-0.2958276867866516, -0.12511926889419556], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001156.pt"}
{"idx": 1158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzrlM6uX9oU.mp4", "duration": 93.72706039372706, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman on a video call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman on a video call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 71.0], "base_se_norm": [0.28031325340270996, 0.5150374174118042], "gt_se_raw": [60.0, 71.0], "gt_se_norm": [0.28031325340270996, 0.5150374174118042], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001158.pt"}
{"idx": 1160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qtLVMNUqlLk.mp4", "duration": 189.314125, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men talking with a city and water in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men talking with a city and water in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.7570176124572754], "gt_se_raw": [0.0, 23.0], "gt_se_norm": [-1.0, -0.7570176124572754], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001160.pt"}
{"idx": 1162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qtLVMNUqlLk.mp4", "duration": 189.314125, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the man in the red shirt's self-introduction?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the man in the red shirt's self-introduction?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 50.0], "base_se_norm": [-0.5879863500595093, -0.47177737951278687], "gt_se_raw": [38.0, 50.0], "gt_se_norm": [-0.5985507965087891, -0.47177737951278687], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001162.pt"}
{"idx": 1164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qtLVMNUqlLk.mp4", "duration": 189.314125, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a speech bubble containing the text 'Publisher's Website' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a speech bubble containing the text 'Publisher's Website' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 156.0], "base_se_norm": [0.6057968139648438, 0.6480545997619629], "gt_se_raw": [152.0, 156.0], "gt_se_norm": [0.6057968139648438, 0.6480545997619629], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001164.pt"}
{"idx": 1166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tByqpujET9o.mp4", "duration": 394.0603333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the women are gathered around the purple dress, and one of them checks the price tag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the women are gathered around the purple dress, and one of them checks the price tag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 41.0], "base_se_norm": [-0.8375883102416992, -0.7919100522994995], "gt_se_raw": [32.0, 40.0], "gt_se_norm": [-0.8375883102416992, -0.796985387802124], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001166.pt"}
{"idx": 1168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tByqpujET9o.mp4", "duration": 394.0603333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where two women are in the bathroom, one vomiting into the toilet and the other sitting on the sink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 247.0 - 260.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where two women are in the bathroom, one vomiting into the toilet and the other sitting on the sink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 247.0 - 260.0 seconds.<|im_end|>\n", "base_se_raw": [247.0, 260.0], "base_se_norm": [0.253615140914917, 0.3195948600769043], "gt_se_raw": [247.0, 259.0], "gt_se_norm": [0.253615140914917, 0.3145195245742798], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001168.pt"}
{"idx": 1170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9iQX13HWUzk.mp4", "duration": 81.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the dog first starts its long, apologetic howl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the dog first starts its long, apologetic howl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 34.0], "base_se_norm": [-0.45734840631484985, -0.1613566279411316], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.6793422698974609, -0.5806783437728882], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001170.pt"}
{"idx": 1172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9iQX13HWUzk.mp4", "duration": 81.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the dog yawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog yawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [0.25796496868133545, 0.3319629430770874], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [0.2826310396194458, 0.3319629430770874], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001172.pt"}
{"idx": 1174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9iQX13HWUzk.mp4", "duration": 81.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person's foot entering the scene from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person's foot entering the scene from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 48.0], "base_se_norm": [0.1346350908279419, 0.18396711349487305], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.1346350908279419, 0.18396711349487305], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001174.pt"}
{"idx": 1176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hm_HXH2z-2M.mp4", "duration": 5.638966666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child touch the side of their head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child touch the side of their head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, 0.06402480602264404], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.6453250646591187, 0.06402480602264404], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001176.pt"}
{"idx": 1178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hm_HXH2z-2M.mp4", "duration": 5.638966666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the child looks up at the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the child looks up at the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.6453250646591187, 0.06402480602264404], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [0.06402480602264404, 0.41869962215423584], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001178.pt"}
{"idx": 1180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SxZgTzn64ME.mp4", "duration": 166.41625, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man is riding as a passenger in a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man is riding as a passenger in a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.795693039894104, -0.7596388459205627], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.795693039894104, -0.7596388459205627], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001180.pt"}
{"idx": 1182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SxZgTzn64ME.mp4", "duration": 166.41625, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where several people are eating hot pot together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 152.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where several people are eating hot pot together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 152.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 152.0], "base_se_norm": [0.718582272529602, 0.826744794845581], "gt_se_raw": [143.0, 152.0], "gt_se_norm": [0.718582272529602, 0.826744794845581], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001182.pt"}
{"idx": 1184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SxZgTzn64ME.mp4", "duration": 166.41625, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man is working on his laptop while sitting on the bleachers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man is working on his laptop while sitting on the bleachers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 100.0], "base_se_norm": [0.11767935752868652, 0.2018057107925415], "gt_se_raw": [92.0, 100.0], "gt_se_norm": [0.10566127300262451, 0.2018057107925415], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001184.pt"}
{"idx": 1186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PbrPhIn0cKk.mp4", "duration": 128.4283, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the man in the suit is talking in the meeting room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the man in the suit is talking in the meeting room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 41.0], "base_se_norm": [-0.5328132510185242, -0.3615114092826843], "gt_se_raw": [29.0, 42.0], "gt_se_norm": [-0.5483860969543457, -0.3459385633468628], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001186.pt"}
{"idx": 1188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PbrPhIn0cKk.mp4", "duration": 128.4283, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the doctors are wearing surgical masks and scrubs in the operating room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the doctors are wearing surgical masks and scrubs in the operating room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 94.0], "base_se_norm": [0.417133092880249, 0.46385180950164795], "gt_se_raw": [91.0, 94.0], "gt_se_norm": [0.417133092880249, 0.46385180950164795], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001188.pt"}
{"idx": 1190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_-O6Ppkrf98.mp4", "duration": 61.2, "prompt_text": "Please find the visual event described by the sentence 'When are the two men crossing the river on the stone bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men crossing the river on the stone bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.9346405267715454, -0.8692810535430908], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9346405267715454, -0.8692810535430908], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001190.pt"}
{"idx": 1192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_-O6Ppkrf98.mp4", "duration": 61.2, "prompt_text": "Please find the visual event described by the sentence 'When does the man shoot an arrow with a rope attached to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man shoot an arrow with a rope attached to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [0.07843136787414551, 0.17647051811218262], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [0.11111104488372803, 0.17647051811218262], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001192.pt"}
{"idx": 1194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_-O6Ppkrf98.mp4", "duration": 61.2, "prompt_text": "Please find the visual event described by the sentence 'When is the castle shown in the rain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the castle shown in the rain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 60.0], "base_se_norm": [0.9281045198440552, 0.9607843160629272], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [0.9281045198440552, 0.9934639930725098], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001194.pt"}
{"idx": 1196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5n0EFtOuH9s.mp4", "duration": 179.77960000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the process of sifting cocoa powder into the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of sifting cocoa powder into the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 53.0], "base_se_norm": [-0.5438859462738037, -0.4103891849517822], "gt_se_raw": [41.0, 53.0], "gt_se_norm": [-0.5438859462738037, -0.4103891849517822], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001196.pt"}
{"idx": 1198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5n0EFtOuH9s.mp4", "duration": 179.77960000000002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the cooled chocolate mixture is shaped into small balls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the cooled chocolate mixture is shaped into small balls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 105.0], "base_se_norm": [0.02347540855407715, 0.1680968999862671], "gt_se_raw": [93.0, 99.0], "gt_se_norm": [0.034600138664245605, 0.10134851932525635], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001198.pt"}
{"idx": 1200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFQB8hJdLZw.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a physical therapist assists a woman who is sitting on a green exercise ball and using a resistance band.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a physical therapist assists a woman who is sitting on a green exercise ball and using a resistance band.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [79, 82], "end_token_span": [84, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001200.pt"}
{"idx": 1202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFQB8hJdLZw.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the time when an instructor is teaching a student about the human rib cage using a skeleton model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when an instructor is teaching a student about the human rib cage using a skeleton model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001202.pt"}
{"idx": 1204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFQB8hJdLZw.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the text 'CAREERS' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the text 'CAREERS' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.41042566299438477, -0.08288431167602539], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001204.pt"}
{"idx": 1206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3LrBM_X-CYc.mp4", "duration": 23.967, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a blue shirt joins the scene from the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a blue shirt joins the scene from the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [0.5020654201507568, 0.8358577489852905], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [0.5020654201507568, 0.7524095773696899], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001206.pt"}
{"idx": 1208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3LrBM_X-CYc.mp4", "duration": 23.967, "prompt_text": "Please find the visual event described by the sentence 'At what time does the boy in the plaid shirt kneel on the floor to gather balloons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the boy in the plaid shirt kneel on the floor to gather balloons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [0.3351691961288452, 0.8358577489852905], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.6689615249633789, 0.9193058013916016], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001208.pt"}
{"idx": 1210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t7Vtyztl4ro.mp4", "duration": 34.134100000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman pick up the bar from the floor for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman pick up the bar from the floor for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.41407567262649536, -0.2382984161376953], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.7070378065109253, -0.5898529887199402], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001210.pt"}
{"idx": 1212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t7Vtyztl4ro.mp4", "duration": 34.134100000000004, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the woman says 'Good job, man'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the woman says 'Good job, man'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.6991804838180542, 0.8163653612136841], "gt_se_raw": [17.0, 18.0], "gt_se_norm": [-0.003928661346435547, 0.054663777351379395], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001212.pt"}
{"idx": 1214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t7Vtyztl4ro.mp4", "duration": 34.134100000000004, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman gives the boy a high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman gives the boy a high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.6991804838180542, 0.8163653612136841], "gt_se_raw": [30.0, 31.0], "gt_se_norm": [0.7577729225158691, 0.8163653612136841], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001214.pt"}
{"idx": 1216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bMz-xG9gk8k.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the teacher guides the student to arrange symbols on the small board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the teacher guides the student to arrange symbols on the small board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 90.0], "base_se_norm": [-0.46719950437545776, 0.49850142002105713], "gt_se_raw": [32.0, 110.0], "gt_se_norm": [-0.46719950437545776, 0.831501841545105], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001216.pt"}
{"idx": 1218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bMz-xG9gk8k.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'At what point are they discussing the 'music with Pascal' activity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are they discussing the 'music with Pascal' activity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 58.0], "base_se_norm": [-0.46719950437545776, -0.03429907560348511], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.24875116348266602, 0.29870128631591797], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001218.pt"}
{"idx": 1220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CSlsvxgfhE.mp4", "duration": 439.50573333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the animated man eating from a can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the animated man eating from a can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 37.0], "base_se_norm": [-0.9726966023445129, -0.8316290378570557], "gt_se_raw": [6.0, 37.0], "gt_se_norm": [-0.9726966023445129, -0.8316290378570557], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001220.pt"}
{"idx": 1222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CSlsvxgfhE.mp4", "duration": 439.50573333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the glowing, spinning cube.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 219.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the glowing, spinning cube.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 219.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 219.0], "base_se_norm": [-0.21275204420089722, -0.0034259557723999023], "gt_se_raw": [172.0, 218.0], "gt_se_norm": [-0.2173026204109192, -0.007976531982421875], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001222.pt"}
{"idx": 1224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CSlsvxgfhE.mp4", "duration": 439.50573333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the scene with multiple clones of the man standing in rows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 383.0 - 409.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene with multiple clones of the man standing in rows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 383.0 - 409.0 seconds.<|im_end|>\n", "base_se_raw": [383.0, 409.0], "base_se_norm": [0.7428668737411499, 0.8611816167831421], "gt_se_raw": [383.0, 408.0], "gt_se_norm": [0.7428668737411499, 0.8566310405731201], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001224.pt"}
{"idx": 1226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdZMypElBpo.mp4", "duration": 61.60320833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the child attempt to use his power on the sleeping dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child attempt to use his power on the sleeping dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.7078074216842651, -0.5454782247543335], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.6753416061401367, -0.5454782247543335], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001226.pt"}
{"idx": 1228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdZMypElBpo.mp4", "duration": 61.60320833333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the car's engine starts and its headlights flash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the car's engine starts and its headlights flash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [0.4609628915786743, 0.5583604574203491], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.49342870712280273, 0.5583604574203491], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001228.pt"}
{"idx": 1230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gq0jfzgvvrM.mp4", "duration": 146.51303333333334, "prompt_text": "Please find the visual event described by the sentence 'When do the little birds laugh at the big bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the little birds laugh at the big bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 55.0], "base_se_norm": [-0.3584188222885132, -0.24921351671218872], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.7679387331008911, -0.6996854543685913], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001230.pt"}
{"idx": 1232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gq0jfzgvvrM.mp4", "duration": 146.51303333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the small birds get squished together in the middle of the wire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the small birds get squished together in the middle of the wire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.3584188222885132, -0.23556286096572876], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.3447681665420532, -0.3174668550491333], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001232.pt"}
{"idx": 1234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gq0jfzgvvrM.mp4", "duration": 146.51303333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the small birds' feathers plucked out as they are launched into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the small birds' feathers plucked out as they are launched into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 102.0], "base_se_norm": [0.35141563415527344, 0.3923676013946533], "gt_se_raw": [99.0, 102.0], "gt_se_norm": [0.35141563415527344, 0.3923676013946533], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001234.pt"}
{"idx": 1236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WZ89Bgcdr1M.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when an instructor is teaching in a classroom setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when an instructor is teaching in a classroom setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001236.pt"}
{"idx": 1238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WZ89Bgcdr1M.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man is shown working closely with a red hose and safety glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man is shown working closely with a red hose and safety glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001238.pt"}
{"idx": 1240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bLX3hv7cTOE.mp4", "duration": 154.4543, "prompt_text": "Please find the visual event described by the sentence 'When is the presenter showing the classic Webster-Pak from the front?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the presenter showing the classic Webster-Pak from the front?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 33.0], "base_se_norm": [-0.7798701524734497, -0.5726891756057739], "gt_se_raw": [34.0, 50.0], "gt_se_norm": [-0.5597403049468994, -0.35255932807922363], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001240.pt"}
{"idx": 1242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bLX3hv7cTOE.mp4", "duration": 154.4543, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the back of the medication pack is being displayed and explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the back of the medication pack is being displayed and explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 60.0], "base_se_norm": [-0.35255932807922363, -0.2230711579322815], "gt_se_raw": [50.0, 60.0], "gt_se_norm": [-0.35255932807922363, -0.2230711579322815], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001242.pt"}
{"idx": 1244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bLX3hv7cTOE.mp4", "duration": 154.4543, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the presenter introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the presenter introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 16.0], "base_se_norm": [-0.9611535668373108, -0.7928189635276794], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9611535668373108, -0.9352559447288513], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001244.pt"}
{"idx": 1246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z0yFqQWLa0M.mp4", "duration": 84.37595555555556, "prompt_text": "Please find the visual event described by the sentence 'At what point do they both agree that 'it's the house'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they both agree that 'it's the house'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [0.06665456295013428, 0.13776493072509766], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.17037975788116455, -0.09926944971084595], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001246.pt"}
{"idx": 1248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z0yFqQWLa0M.mp4", "duration": 84.37595555555556, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman walks away and opens the door as if to exit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman walks away and opens the door as if to exit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [0.5644267797470093, 0.6829439401626587], "gt_se_raw": [67.0, 70.0], "gt_se_norm": [0.5881302356719971, 0.6592404842376709], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001248.pt"}
{"idx": 1250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/db3Ep-jM6ZE.mp4", "duration": 17.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the person first appear and walk towards the counter on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person first appear and walk towards the counter on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.8865784406661987, -0.2060491442680359], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.8865784406661987, -0.5463137626647949], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001250.pt"}
{"idx": 1252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/db3Ep-jM6ZE.mp4", "duration": 17.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman seen wiping the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman seen wiping the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.2060491442680359, 0.5879017114639282], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.09262758493423462, 0.47448015213012695], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001252.pt"}
{"idx": 1254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/db3Ep-jM6ZE.mp4", "duration": 17.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the total duration the person is visible in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the total duration the person is visible in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 15.0], "base_se_norm": [-0.8865784406661987, 0.7013232707977295], "gt_se_raw": [1.0, 16.0], "gt_se_norm": [-0.8865784406661987, 0.8147448301315308], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001254.pt"}
{"idx": 1256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jMsTvH6gTGk.mp4", "duration": 184.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video introduce the concept of an AB pattern using buttons and pegs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video introduce the concept of an AB pattern using buttons and pegs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 47.0], "base_se_norm": [-0.7291440963745117, -0.4907909035682678], "gt_se_raw": [25.0, 48.0], "gt_se_norm": [-0.7291440963745117, -0.4799566864967346], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001256.pt"}
{"idx": 1258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jMsTvH6gTGk.mp4", "duration": 184.6, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the video changes the pattern from ABC to ABB.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the video changes the pattern from ABC to ABB.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 114.0], "base_se_norm": [0.09425783157348633, 0.23510289192199707], "gt_se_raw": [101.0, 112.0], "gt_se_norm": [0.09425783157348633, 0.21343445777893066], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001258.pt"}
{"idx": 1260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QNJx7Vi4Sg4.mp4", "duration": 133.80033333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in the pink swimsuit speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in the pink swimsuit speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 32.0], "base_se_norm": [-0.8804188370704651, -0.5216753482818604], "gt_se_raw": [8.0, 32.0], "gt_se_norm": [-0.8804188370704651, -0.5216753482818604], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001260.pt"}
{"idx": 1262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QNJx7Vi4Sg4.mp4", "duration": 133.80033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman covers her mouth in surprise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman covers her mouth in surprise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [-0.1330365538597107, -0.07324600219726562], "gt_se_raw": [59.0, 69.0], "gt_se_norm": [-0.11808890104293823, 0.0313875675201416], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001262.pt"}
{"idx": 1264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QNJx7Vi4Sg4.mp4", "duration": 133.80033333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman and the man kiss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman and the man kiss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 88.0], "base_se_norm": [0.04633522033691406, 0.31539273262023926], "gt_se_raw": [119.0, 122.0], "gt_se_norm": [0.7787697315216064, 0.8236126899719238], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001264.pt"}
{"idx": 1266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vm1Zeu-DpmU.mp4", "duration": 201.70149999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the animation of a woman standing next to a trash can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the animation of a woman standing next to a trash can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 62.0], "base_se_norm": [-0.5141335129737854, -0.3852301836013794], "gt_se_raw": [48.0, 61.0], "gt_se_norm": [-0.5240491628646851, -0.39514583349227905], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001266.pt"}
{"idx": 1268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vm1Zeu-DpmU.mp4", "duration": 201.70149999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see the marketing manager speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the marketing manager speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 156.0], "base_se_norm": [0.4179368019104004, 0.5468401908874512], "gt_se_raw": [143.0, 155.0], "gt_se_norm": [0.4179368019104004, 0.5369246006011963], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001268.pt"}
{"idx": 1270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OpcxenhBO64.mp4", "duration": 195.2951, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the paper being folded in half diagonally for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the paper being folded in half diagonally for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.9795181751251221, -0.8873499631881714], "gt_se_raw": [2.0, 11.0], "gt_se_norm": [-0.9795181751251221, -0.8873499631881714], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001270.pt"}
{"idx": 1272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OpcxenhBO64.mp4", "duration": 195.2951, "prompt_text": "Please find the visual event described by the sentence 'Show me when the side corners are folded inwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the side corners are folded inwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 84.0], "base_se_norm": [-0.3343406915664673, -0.13976341485977173], "gt_se_raw": [50.0, 70.0], "gt_se_norm": [-0.48795437812805176, -0.28313612937927246], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001272.pt"}
{"idx": 1274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OpcxenhBO64.mp4", "duration": 195.2951, "prompt_text": "Please find the visual event described by the sentence 'Show me when scissors are used to cut a notch in the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when scissors are used to cut a notch in the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 169.0], "base_se_norm": [0.6795096397399902, 0.7307140827178955], "gt_se_raw": [164.0, 169.0], "gt_se_norm": [0.6795096397399902, 0.7307140827178955], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001274.pt"}
{"idx": 1276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPbvr9LH84w.mp4", "duration": 184.184, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in the yellow cardigan has a seizure and collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in the yellow cardigan has a seizure and collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [-0.2616080045700073, -0.20731443166732788], "gt_se_raw": [67.0, 73.0], "gt_se_norm": [-0.27246665954589844, -0.20731443166732788], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001276.pt"}
{"idx": 1278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPbvr9LH84w.mp4", "duration": 184.184, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the man with red hair is making a call from a payphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the man with red hair is making a call from a payphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 114.0], "base_se_norm": [0.12930548191070557, 0.2378925085067749], "gt_se_raw": [105.0, 114.0], "gt_se_norm": [0.14016413688659668, 0.2378925085067749], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001278.pt"}
{"idx": 1280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9jX0S7mvB8.mp4", "duration": 151.6515, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men shaking hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men shaking hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.8813068270683289], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9736237525939941, -0.9472475051879883], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001280.pt"}
{"idx": 1282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9jX0S7mvB8.mp4", "duration": 151.6515, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the vintage car drives by.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the vintage car drives by.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [-0.5911679267883301, -0.5252272486686707], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.5516035556793213, -0.5252272486686707], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001282.pt"}
{"idx": 1284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9jX0S7mvB8.mp4", "duration": 151.6515, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the woman takes off her shawl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the woman takes off her shawl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6439204216003418, -0.604356050491333], "gt_se_raw": [115.0, 117.0], "gt_se_norm": [0.5166351795196533, 0.5430114269256592], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001284.pt"}
{"idx": 1286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-wHytb5Fe2k.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the boy selects the final letter, 'Z', to complete his task.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the boy selects the final letter, 'Z', to complete his task.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [0.3457944393157959, 0.4953271150588989], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [-0.47663551568984985, -0.2897196412086487], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001286.pt"}
{"idx": 1288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-wHytb5Fe2k.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy uses the communication device to say 'I want the spinning toy'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy uses the communication device to say 'I want the spinning toy'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 41.0], "base_se_norm": [-0.06542056798934937, 0.5327103137969971], "gt_se_raw": [32.0, 42.0], "gt_se_norm": [0.19626164436340332, 0.5700935125350952], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001288.pt"}
{"idx": 1290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uWuOBonC1ds.mp4", "duration": 312.37873333333334, "prompt_text": "Please find the visual event described by the sentence 'When do the children get their shopping lists from the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the children get their shopping lists from the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 28.0], "base_se_norm": [-0.8783527612686157, -0.820730447769165], "gt_se_raw": [19.0, 27.0], "gt_se_norm": [-0.8783527612686157, -0.8271329402923584], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001290.pt"}
{"idx": 1292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uWuOBonC1ds.mp4", "duration": 312.37873333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy is checking 'cookies' off his list with a green marker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy is checking 'cookies' off his list with a green marker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 189.0], "base_se_norm": [0.13323974609375, 0.2100696563720703], "gt_se_raw": [183.0, 189.0], "gt_se_norm": [0.17165470123291016, 0.2100696563720703], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001292.pt"}
{"idx": 1294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uWuOBonC1ds.mp4", "duration": 312.37873333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the boy pick up a container of tomatoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 261.0 - 265.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy pick up a container of tomatoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 261.0 - 265.0 seconds.<|im_end|>\n", "base_se_raw": [261.0, 265.0], "base_se_norm": [0.6710485219955444, 0.6966584920883179], "gt_se_raw": [261.0, 264.0], "gt_se_norm": [0.6710485219955444, 0.6902559995651245], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001294.pt"}
{"idx": 1296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PG47b8hd2o.mp4", "duration": 85.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the speaker suggests that embracing change can be difficult.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the speaker suggests that embracing change can be difficult.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [0.03004288673400879, 0.17050325870513916], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.1338275671005249, -0.08700740337371826], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001296.pt"}
{"idx": 1298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PG47b8hd2o.mp4", "duration": 85.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where he says that the viewer's vision for the course needs to be solidified.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where he says that the viewer's vision for the course needs to be solidified.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 56.0], "base_se_norm": [0.17050325870513916, 0.3109637498855591], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [0.33437371253967285, 0.4046039581298828], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001298.pt"}
{"idx": 1300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Nb8KXFH3aMs.mp4", "duration": 54.68, "prompt_text": "Please find the visual event described by the sentence 'When can we see two dogs playing under the trees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two dogs playing under the trees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8536942005157471], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8536942005157471], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001300.pt"}
{"idx": 1302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Nb8KXFH3aMs.mp4", "duration": 54.68, "prompt_text": "Please find the visual event described by the sentence 'At what point is a close-up of the yellow sign shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a close-up of the yellow sign shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.5245062112808228, -0.34162402153015137], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.41477686166763306, -0.34162402153015137], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001302.pt"}
{"idx": 1304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Nb8KXFH3aMs.mp4", "duration": 54.68, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man picks up after his dog and throws the waste in the bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man picks up after his dog and throws the waste in the bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [0.42648136615753174, 0.5727871656417847], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [0.38990485668182373, 0.5727871656417847], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001304.pt"}
{"idx": 1306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgg9jE1jXjg.mp4", "duration": 161.69486666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a marching band playing their instruments.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a marching band playing their instruments.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 44.0], "base_se_norm": [-0.5176099538803101, -0.4557650685310364], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.5052409172058105, -0.4557650685310364], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001306.pt"}
{"idx": 1308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgg9jE1jXjg.mp4", "duration": 161.69486666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a coach sign a football?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a coach sign a football?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 51.0], "base_se_norm": [-0.40628916025161743, -0.3691822290420532], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.40628916025161743, -0.3691822290420532], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001308.pt"}
{"idx": 1310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zfn8Mtyu5s0.mp4", "duration": 113.9, "prompt_text": "Please find the visual event described by the sentence 'At what point is the hank of yarn being untwisted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the hank of yarn being untwisted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 23.0], "base_se_norm": [-0.771729588508606, -0.5961369276046753], "gt_se_raw": [12.0, 18.0], "gt_se_norm": [-0.7892888784408569, -0.6839332580566406], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001310.pt"}
{"idx": 1312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zfn8Mtyu5s0.mp4", "duration": 113.9, "prompt_text": "Please find the visual event described by the sentence 'When does the winding start to go at an angle over the initial loops?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the winding start to go at an angle over the initial loops?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 62.0], "base_se_norm": [-0.13959616422653198, 0.08867430686950684], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.31518876552581787, -0.24495172500610352], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001312.pt"}
{"idx": 1314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zfn8Mtyu5s0.mp4", "duration": 113.9, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the nostepinne is removed from the finished yarn ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the nostepinne is removed from the finished yarn ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 104.0], "base_se_norm": [0.773485541343689, 0.8261632919311523], "gt_se_raw": [101.0, 104.0], "gt_se_norm": [0.773485541343689, 0.8261632919311523], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001314.pt"}
{"idx": 1316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSfeGBrchjU.mp4", "duration": 110.56, "prompt_text": "Please find the visual event described by the sentence 'At what point does a girl in a light blue sweater get pulled out of the group?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a girl in a light blue sweater get pulled out of the group?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 68.0], "base_se_norm": [0.12156295776367188, 0.2301013469696045], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [0.13965272903442383, 0.2301013469696045], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001316.pt"}
{"idx": 1318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSfeGBrchjU.mp4", "duration": 110.56, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a boy in a red shirt smiles at the camera while running.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a boy in a red shirt smiles at the camera while running.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.5477568507194519, -0.4753979444503784], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.6743849515914917, -0.6382055282592773], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001318.pt"}
{"idx": 1320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3nYvJY2C9B4.mp4", "duration": 86.01926666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the cat walk on the blue sofa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat walk on the blue sofa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 42.0], "base_se_norm": [-0.906997561454773, -0.023474574089050293], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.906997561454773, -0.7907445430755615], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001320.pt"}
{"idx": 1322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3nYvJY2C9B4.mp4", "duration": 86.01926666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person pick up the cat to give it the yogurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person pick up the cat to give it the yogurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [-0.023474574089050293, 0.11602902412414551], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [-0.04672515392303467, 0.04627728462219238], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001322.pt"}
{"idx": 1324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3nYvJY2C9B4.mp4", "duration": 86.01926666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the cat have the yogurt cup stuck on its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat have the yogurt cup stuck on its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 81.0], "base_se_norm": [0.04627728462219238, 0.8832991123199463], "gt_se_raw": [72.0, 76.0], "gt_se_norm": [0.6740436553955078, 0.7670460939407349], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001324.pt"}
{"idx": 1326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ehJfUqSbPtg.mp4", "duration": 88.09693031441432, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the cat on the left is startled by the robot vacuum and jumps back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the cat on the left is startled by the robot vacuum and jumps back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 56.0], "base_se_norm": [0.08970880508422852, 0.2713268995285034], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8410841226577759, -0.7956795692443848], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001326.pt"}
{"idx": 1328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ehJfUqSbPtg.mp4", "duration": 88.09693031441432, "prompt_text": "Please find the visual event described by the sentence 'At what point do two cats walk into the hallway on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two cats walk into the hallway on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.5913591980934143, -0.4097411036491394], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.1124110221862793, 0.1805177927017212], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001328.pt"}
{"idx": 1330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIb_741_dIw.mp4", "duration": 89.214125, "prompt_text": "Please find the visual event described by the sentence 'Show me when the chocolate bar is being chopped.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the chocolate bar is being chopped.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9327460527420044, -0.8430742025375366], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9327460527420044, -0.8430742025375366], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001330.pt"}
{"idx": 1332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIb_741_dIw.mp4", "duration": 89.214125, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the dry ingredients are sifted into the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the dry ingredients are sifted into the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.34987872838974, -0.14811694622039795], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.37229669094085693, -0.26020681858062744], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001332.pt"}
{"idx": 1334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIb_741_dIw.mp4", "duration": 89.214125, "prompt_text": "Please find the visual event described by the sentence 'When is the finished brownie being cut into individual squares?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the finished brownie being cut into individual squares?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 67.0], "base_se_norm": [0.3450784683227539, 0.5020042657852173], "gt_se_raw": [60.0, 67.0], "gt_se_norm": [0.3450784683227539, 0.5020042657852173], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001334.pt"}
{"idx": 1336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJnpwL-2Drc.mp4", "duration": 142.28, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where someone is sentenced to jazz.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where someone is sentenced to jazz.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.5501827001571655, -0.4939555525779724], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.564239501953125, -0.5080123543739319], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001336.pt"}
{"idx": 1338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJnpwL-2Drc.mp4", "duration": 142.28, "prompt_text": "Please find the visual event described by the sentence 'Where can I find the 'pull my finger' joke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find the 'pull my finger' joke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 41.0], "base_se_norm": [-0.4939555525779724, -0.4236716032028198], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.4798988103866577, -0.4236716032028198], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001338.pt"}
{"idx": 1340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VGb1wgIkPEw.mp4", "duration": 11.541666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the woman seen standing next to the maroon SUV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman seen standing next to the maroon SUV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.4801444411277771], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.30685919523239136], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001340.pt"}
{"idx": 1342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VGb1wgIkPEw.mp4", "duration": 11.541666666666666, "prompt_text": "Please find the visual event described by the sentence 'During which interval is the sound of a loud siren audible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which interval is the sound of a loud siren audible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.30685919523239136, 0.7328519821166992], "gt_se_raw": [1.0, 11.0], "gt_se_norm": [-0.826714813709259, 0.9061371088027954], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001342.pt"}
{"idx": 1344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VGb1wgIkPEw.mp4", "duration": 11.541666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can a white semi-truck be seen driving on the distant road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can a white semi-truck be seen driving on the distant road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.4801444411277771], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.30685919523239136], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001344.pt"}
{"idx": 1346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkTpA2U4ooA.mp4", "duration": 150.98416666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the player in yellow armor build a bridge towards the green team's island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in yellow armor build a bridge towards the green team's island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.6158537864685059, -0.5496216416358948], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.6158537864685059, -0.5761144757270813], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001346.pt"}
{"idx": 1348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkTpA2U4ooA.mp4", "duration": 150.98416666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the player pour water inside the blue team's base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the player pour water inside the blue team's base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 77.0], "base_se_norm": [-0.08599686622619629, 0.019974589347839355], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [-0.05950397253036499, -0.03301113843917847], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001348.pt"}
{"idx": 1350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-tH4HbncWNs.mp4", "duration": 122.49737499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the Tigger Tsum Tsum turn into multiple Tiggers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Tigger Tsum Tsum turn into multiple Tiggers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [-0.3632516860961914, -0.3142710328102112], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.3469247817993164, -0.3142710328102112], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001350.pt"}
{"idx": 1352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-tH4HbncWNs.mp4", "duration": 122.49737499999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a strawberry is launched into the air by a spoon seesaw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a strawberry is launched into the air by a spoon seesaw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 70.0], "base_se_norm": [0.09390103816986084, 0.1428816318511963], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [0.11022782325744629, 0.1428816318511963], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001352.pt"}
{"idx": 1354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-tH4HbncWNs.mp4", "duration": 122.49737499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where Mickey places a strawberry on top of the cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where Mickey places a strawberry on top of the cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [-0.5428473353385925, -0.4938666820526123], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [-0.5428473353385925, -0.4938666820526123], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001354.pt"}
{"idx": 1356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/beECZjCRLmQ.mp4", "duration": 33.44, "prompt_text": "Please find the visual event described by the sentence 'Show the action of taking the pepper pot out of its packaging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the action of taking the pepper pot out of its packaging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.7009569406509399, -0.4617224931716919], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.6411483287811279, -0.4617224931716919], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001356.pt"}
{"idx": 1358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/beECZjCRLmQ.mp4", "duration": 33.44, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when both the salt and pepper pots are placed side-by-side on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when both the salt and pepper pots are placed side-by-side on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 22.0], "base_se_norm": [-0.22248798608779907, 0.3157895803451538], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.3421052098274231, 0.07655501365661621], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001358.pt"}
{"idx": 1360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QmJ3xuJrUcM.mp4", "duration": 146.47966666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the man explaining the three-step process of the experiment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man explaining the three-step process of the experiment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 45.0], "base_se_norm": [-0.8634622693061829, -0.38558024168014526], "gt_se_raw": [10.0, 23.0], "gt_se_norm": [-0.8634622693061829, -0.6859632134437561], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001360.pt"}
{"idx": 1362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QmJ3xuJrUcM.mp4", "duration": 146.47966666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the crow pull the string with its beak and feet to get the short stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the crow pull the string with its beak and feet to get the short stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 103.0], "base_se_norm": [0.14691686630249023, 0.40633857250213623], "gt_se_raw": [85.0, 102.0], "gt_se_norm": [0.16057074069976807, 0.39268481731414795], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001362.pt"}
{"idx": 1364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QmJ3xuJrUcM.mp4", "duration": 146.47966666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the crow use the long stick to retrieve the food from the transparent box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the crow use the long stick to retrieve the food from the transparent box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 138.0], "base_se_norm": [0.7613366842269897, 0.8842206001281738], "gt_se_raw": [127.0, 138.0], "gt_se_norm": [0.7340291738510132, 0.8842206001281738], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001364.pt"}
{"idx": 1366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zMB-bVwEi2Q.mp4", "duration": 72.33893333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy types the letter 'a'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy types the letter 'a'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.19821876287460327, 0.05060982704162598], "gt_se_raw": [29.0, 30.0], "gt_se_norm": [-0.19821876287460327, -0.17057114839553833], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001366.pt"}
{"idx": 1368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zMB-bVwEi2Q.mp4", "duration": 72.33893333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the boy is typing very fast and continuously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the boy is typing very fast and continuously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 39.0], "base_se_norm": [-0.5852855443954468, 0.07825744152069092], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.797095775604248, 0.8800387382507324], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001368.pt"}
{"idx": 1370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mTLO2F_ERY.mp4", "duration": 123.84, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man flipping an umbrella with wind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man flipping an umbrella with wind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.24095606803894043, -0.19250643253326416], "gt_se_raw": [45.0, 50.0], "gt_se_norm": [-0.27325576543807983, -0.19250643253326416], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001370.pt"}
{"idx": 1372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mTLO2F_ERY.mp4", "duration": 123.84, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man is blowing a toy windmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man is blowing a toy windmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 115.0], "base_se_norm": [0.776485800743103, 0.8572351932525635], "gt_se_raw": [111.0, 116.0], "gt_se_norm": [0.7926356792449951, 0.8733850717544556], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001372.pt"}
{"idx": 1374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R93ce4FZGbc.mp4", "duration": 116.116, "prompt_text": "Please find the visual event described by the sentence 'When does the Daddy Shark sing its part of the song?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Daddy Shark sing its part of the song?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 52.0], "base_se_norm": [-0.2421371340751648, -0.10434389114379883], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [-0.22491300106048584, -0.10434389114379883], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001374.pt"}
{"idx": 1376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R93ce4FZGbc.mp4", "duration": 116.116, "prompt_text": "Please find the visual event described by the sentence 'When are the fish hiding inside the rock, safe from the sharks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the fish hiding inside the rock, safe from the sharks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 95.0], "base_se_norm": [0.48127734661102295, 0.6362947225570679], "gt_se_raw": [86.0, 95.0], "gt_se_norm": [0.48127734661102295, 0.6362947225570679], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001376.pt"}
{"idx": 1378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R93ce4FZGbc.mp4", "duration": 116.116, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the yellow Baby Shark is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the yellow Baby Shark is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [-0.5693961381912231, -0.3971545696258545], "gt_se_raw": [27.0, 35.0], "gt_se_norm": [-0.5349478125572205, -0.3971545696258545], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001378.pt"}
{"idx": 1380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dKXpAC311Lg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator state that the employment rate for people with disabilities is nearly half that of those without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator state that the employment rate for people with disabilities is nearly half that of those without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.030096113681793213, 0.35786545276641846], "start_token_span": [78, 81], "end_token_span": [83, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001380.pt"}
{"idx": 1382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dKXpAC311Lg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the speaker introduces the main finding from the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the speaker introduces the main finding from the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.4180576801300049], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001382.pt"}
{"idx": 1384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKDyFgUqWfQ.mp4", "duration": 224.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the silverfish attacking the player in green?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the silverfish attacking the player in green?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 189.0], "base_se_norm": [0.5794169902801514, 0.6864961385726929], "gt_se_raw": [174.0, 189.0], "gt_se_norm": [0.5526472330093384, 0.6864961385726929], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001384.pt"}
{"idx": 1386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKDyFgUqWfQ.mp4", "duration": 224.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the player in cyan armor is placing obsidian blocks around the bed for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the player in cyan armor is placing obsidian blocks around the bed for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 37.0], "base_se_norm": [-0.732302188873291, -0.6698393821716309], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.7233788967132568, -0.6966091394424438], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001386.pt"}
{"idx": 1388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKDyFgUqWfQ.mp4", "duration": 224.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player in green get their revenge by knocking the player in cyan off the island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in green get their revenge by knocking the player in cyan off the island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 142.0], "base_se_norm": [0.21356332302093506, 0.2671029567718506], "gt_se_raw": [197.0, 203.0], "gt_se_norm": [0.7578822374343872, 0.8114217519760132], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001388.pt"}
{"idx": 1390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pU0GSbe6r_4.mp4", "duration": 145.92, "prompt_text": "Please find the visual event described by the sentence 'At what point does the fly drink the milk and fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the fly drink the milk and fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.7395833730697632, -0.6984648704528809], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.7532894611358643, -0.7121710777282715], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001390.pt"}
{"idx": 1392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pU0GSbe6r_4.mp4", "duration": 145.92, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the muscular Jerry tears a book apart with his bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the muscular Jerry tears a book apart with his bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [0.041666626930236816, 0.11019742488861084], "gt_se_raw": [77.0, 82.0], "gt_se_norm": [0.05537283420562744, 0.12390351295471191], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001392.pt"}
{"idx": 1394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1yJA_vE6t9E.mp4", "duration": 166.69986666666668, "prompt_text": "Please find the visual event described by the sentence 'A man and a woman are communicating through touch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man and a woman are communicating through touch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 34.0], "base_se_norm": [-0.8680262565612793, -0.5920812487602234], "gt_se_raw": [10.0, 35.0], "gt_se_norm": [-0.8800238966941833, -0.5800836086273193], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001394.pt"}
{"idx": 1396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1yJA_vE6t9E.mp4", "duration": 166.69986666666668, "prompt_text": "Please find the visual event described by the sentence 'A girl wearing gloves and a hearing aid is touching a board with symbols on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A girl wearing gloves and a hearing aid is touching a board with symbols on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 118.0], "base_se_norm": [0.3317347764968872, 0.41571807861328125], "gt_se_raw": [111.0, 119.0], "gt_se_norm": [0.3317347764968872, 0.4277156591415405], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001396.pt"}
{"idx": 1398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1yJA_vE6t9E.mp4", "duration": 166.69986666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman guides a boy's hand to feel a toy car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman guides a boy's hand to feel a toy car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 137.0], "base_se_norm": [0.4277156591415405, 0.6436727046966553], "gt_se_raw": [151.0, 157.0], "gt_se_norm": [0.8116391897201538, 0.8836249113082886], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001398.pt"}
{"idx": 1400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpBW-5BtPoU.mp4", "duration": 124.79133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man demonstrates adjusting the pitch of the theremin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man demonstrates adjusting the pitch of the theremin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 45.0], "base_se_norm": [-0.519197404384613, -0.27879607677459717], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.3429030776023865, -0.2948228120803833], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001400.pt"}
{"idx": 1402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpBW-5BtPoU.mp4", "duration": 124.79133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man plays a short musical phrase combining both pitch and volume control.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man plays a short musical phrase combining both pitch and volume control.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 95.0], "base_se_norm": [0.36227405071258545, 0.5225416421890259], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [-0.1666088104248047, -0.07044827938079834], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001402.pt"}
{"idx": 1404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OSKNiwkaRRo.mp4", "duration": 19.5195, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman in black giving a thumbs up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman in black giving a thumbs up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [0.5369246006011963, 0.7418478727340698], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.7950767278671265], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001404.pt"}
{"idx": 1406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OSKNiwkaRRo.mp4", "duration": 19.5195, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the teacher starts speaking and gesturing energetically.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the teacher starts speaking and gesturing energetically.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.2827685475349426, 0.024616360664367676], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.2827685475349426, 0.024616360664367676], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001406.pt"}
{"idx": 1408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OSKNiwkaRRo.mp4", "duration": 19.5195, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the woman in black points to the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the woman in black points to the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.5901534557342529], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [0.33200132846832275, 0.6393861770629883], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001408.pt"}
{"idx": 1410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0P1Td5OTS-A.mp4", "duration": 40.04, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the ability to open multiple windows is demonstrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the ability to open multiple windows is demonstrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.15084916353225708, -0.0009990334510803223], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.1008991003036499, -0.0009990334510803223], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001410.pt"}
{"idx": 1412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0P1Td5OTS-A.mp4", "duration": 40.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show the automatic refresh feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show the automatic refresh feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [0.1988011598587036, 0.34865128993988037], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.1988011598587036, 0.29870128631591797], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001412.pt"}
{"idx": 1414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4A9lG59Tosg.mp4", "duration": 63.60520833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the complete miniature house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the complete miniature house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 27.0], "base_se_norm": [-1.0, -0.15101295709609985], "gt_se_raw": [0.0, 22.0], "gt_se_norm": [-1.0, -0.30823278427124023], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001414.pt"}
{"idx": 1416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4A9lG59Tosg.mp4", "duration": 63.60520833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the hammer is lifted off the table after the impact.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the hammer is lifted off the table after the impact.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.2139008641242981, -0.11956894397735596], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.2139008641242981, -0.11956894397735596], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001416.pt"}
{"idx": 1418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4A9lG59Tosg.mp4", "duration": 63.60520833333333, "prompt_text": "Please find the visual event described by the sentence 'When does a person in khaki pants approach the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a person in khaki pants approach the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 22.0], "base_se_norm": [-0.7170042991638184, -0.30823278427124023], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.7170042991638184, -0.622672438621521], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001418.pt"}
{"idx": 1420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5iu2FhCiQFw.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl touches the object representing a present.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl touches the object representing a present.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 26.0], "base_se_norm": [-0.7583062052726746, -0.5166124105453491], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.5166124105453491, -0.44224512577056885], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001420.pt"}
{"idx": 1422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5iu2FhCiQFw.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the teacher helps the girl touch the object representing a letter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the teacher helps the girl touch the object representing a letter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.49802058935165405, -0.4236533045768738], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.15269339084625244, 0.24565255641937256], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001422.pt"}
{"idx": 1424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qcf0KPEM-y8.mp4", "duration": 8.074733333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the person place the bracelet on the boy's wrist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person place the bracelet on the boy's wrist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.23843097686767578], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.23843097686767578], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001424.pt"}
{"idx": 1426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qcf0KPEM-y8.mp4", "duration": 8.074733333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy is being pulled up by his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy is being pulled up by his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 7.0], "base_se_norm": [0.23843097686767578, 0.7338032722473145], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [0.4861171245574951, 0.9814895391464233], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001426.pt"}
{"idx": 1428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qcf0KPEM-y8.mp4", "duration": 8.074733333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the person's hand is adjusting the white bracelet on the boy's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the person's hand is adjusting the white bracelet on the boy's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.23843097686767578], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.25694143772125244, 0.4861171245574951], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001428.pt"}
{"idx": 1430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99jwJbaMG1Q.mp4", "duration": 69.069, "prompt_text": "Please find the visual event described by the sentence 'Show me when the dog is lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the dog is lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9131303429603577], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9420868754386902], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001430.pt"}
{"idx": 1432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99jwJbaMG1Q.mp4", "duration": 69.069, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the dog jumps on its hind legs towards the chicken for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the dog jumps on its hind legs towards the chicken for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9420868754386902, -0.8552172183990479], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9131303429603577, -0.8552172183990479], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001432.pt"}
{"idx": 1434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LeTwAJ1cRa8.mp4", "duration": 146.71323333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men in wheelchairs sitting and talking with another man at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men in wheelchairs sitting and talking with another man at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 64.0], "base_se_norm": [-0.23660600185394287, -0.12754970788955688], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [-0.23660600185394287, -0.2093418836593628], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001434.pt"}
{"idx": 1436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LeTwAJ1cRa8.mp4", "duration": 146.71323333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman is seen sleeping on her desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman is seen sleeping on her desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 128.0], "base_se_norm": [0.6767404079437256, 0.7449005842208862], "gt_se_raw": [124.0, 127.0], "gt_se_norm": [0.6903724670410156, 0.7312686443328857], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001436.pt"}
{"idx": 1438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LeTwAJ1cRa8.mp4", "duration": 146.71323333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for the scene where a woman assists another woman in stirring a mixture in a green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for the scene where a woman assists another woman in stirring a mixture in a green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.6864631772041321, -0.6455670595169067], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [-0.6864631772041321, -0.6455670595169067], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001438.pt"}
{"idx": 1440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H5MKL1duTrg.mp4", "duration": 196.59640000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a blue 'PLEASE MAINTAIN SOCIAL DISTANCE' sticker is on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a blue 'PLEASE MAINTAIN SOCIAL DISTANCE' sticker is on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 96.0], "base_se_norm": [-0.13528430461883545, -0.023379921913146973], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [-0.13528430461883545, -0.10476493835449219], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001440.pt"}
{"idx": 1442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H5MKL1duTrg.mp4", "duration": 196.59640000000002, "prompt_text": "Please find the visual event described by the sentence 'At what time are the icons for face-to-face, hybrid, and virtual classes displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the icons for face-to-face, hybrid, and virtual classes displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 128.0], "base_se_norm": [0.1800825595855713, 0.3021601438522339], "gt_se_raw": [116.0, 128.0], "gt_se_norm": [0.1800825595855713, 0.3021601438522339], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001442.pt"}
{"idx": 1444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O_HyZ5aW76c.mp4", "duration": 94.8948, "prompt_text": "Please find the visual event described by the sentence 'When does the man say, 'There's no words on it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say, 'There's no words on it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.11480921506881714, -0.030505359172821045], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.641708493232727, -0.599556565284729], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001444.pt"}
{"idx": 1446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O_HyZ5aW76c.mp4", "duration": 94.8948, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man shout 'We'll do it live!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man shout 'We'll do it live!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 70.0], "base_se_norm": [0.4331660270690918, 0.47531795501708984], "gt_se_raw": [64.0, 68.0], "gt_se_norm": [0.3488621711730957, 0.4331660270690918], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001446.pt"}
{"idx": 1448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O_HyZ5aW76c.mp4", "duration": 94.8948, "prompt_text": "Please find the visual event described by the sentence 'When does the man finally manage to say his closing lines correctly, mentioning Sting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man finally manage to say his closing lines correctly, mentioning Sting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 87.0], "base_se_norm": [0.6650017499923706, 0.8336094617843628], "gt_se_raw": [78.0, 86.0], "gt_se_norm": [0.643925666809082, 0.8125334978103638], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001448.pt"}
{"idx": 1450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VwJCHnk7C-Q.mp4", "duration": 60.2, "prompt_text": "Please find the visual event described by the sentence 'Show the clip presenting the Venice Canals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip presenting the Venice Canals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 24.0], "base_se_norm": [-0.5016611218452454, -0.2026578187942505], "gt_se_raw": [15.0, 25.0], "gt_se_norm": [-0.5016611218452454, -0.16943520307540894], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001450.pt"}
{"idx": 1452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VwJCHnk7C-Q.mp4", "duration": 60.2, "prompt_text": "Please find the visual event described by the sentence 'Find the segment about Lake Como.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment about Lake Como.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 44.0], "base_se_norm": [0.16279065608978271, 0.46179401874542236], "gt_se_raw": [35.0, 45.0], "gt_se_norm": [0.16279065608978271, 0.49501657485961914], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001452.pt"}
{"idx": 1454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L-LE-j2zkCU.mp4", "duration": 159.7012, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with purple hair show her laptop screen to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with purple hair show her laptop screen to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 26.0], "base_se_norm": [-0.8371959924697876, -0.6743919253349304], "gt_se_raw": [13.0, 27.0], "gt_se_norm": [-0.8371959924697876, -0.6618685722351074], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001454.pt"}
{"idx": 1456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L-LE-j2zkCU.mp4", "duration": 159.7012, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a woman in a leopard-print top is looking for spoons in a drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a woman in a leopard-print top is looking for spoons in a drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 59.0], "base_se_norm": [-0.2861669063568115, -0.26112014055252075], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [-0.2861669063568115, -0.23607337474822998], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001456.pt"}
{"idx": 1458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L-LE-j2zkCU.mp4", "duration": 159.7012, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman with long brown hair holds up a cup filled with spoons and pens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman with long brown hair holds up a cup filled with spoons and pens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 128.0], "base_se_norm": [0.39009594917297363, 0.6029936075210571], "gt_se_raw": [112.0, 125.0], "gt_se_norm": [0.4026193618774414, 0.5654233694076538], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001458.pt"}
{"idx": 1460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iDnE3PV4YNc.mp4", "duration": 193.27641666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man is run over by the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man is run over by the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.19000554084777832, 0.22104918956756592], "gt_se_raw": [115.0, 116.0], "gt_se_norm": [0.19000554084777832, 0.20035338401794434], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001460.pt"}
{"idx": 1462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iDnE3PV4YNc.mp4", "duration": 193.27641666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the young man jump over the police car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 146.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the young man jump over the police car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 146.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [146.0, 149.0], "base_se_norm": [0.5107896327972412, 0.5418332815170288], "gt_se_raw": [147.0, 148.0], "gt_se_norm": [0.5211374759674072, 0.5314853191375732], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001462.pt"}
{"idx": 1464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TQQXstNh45g.mp4", "duration": 90.0, "prompt_text": "Please find the visual event described by the sentence 'When does the orangutan climb on the plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the orangutan climb on the plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.800000011920929, -0.7333333492279053], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.800000011920929, -0.7333333492279053], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001464.pt"}
{"idx": 1466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TQQXstNh45g.mp4", "duration": 90.0, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a machine is destroying the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a machine is destroying the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 54.0], "base_se_norm": [-0.2888888716697693, 0.20000004768371582], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.2222222089767456, -0.15555554628372192], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001466.pt"}
{"idx": 1468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TQQXstNh45g.mp4", "duration": 90.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the girl gives the orangutan a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the girl gives the orangutan a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 80.0], "base_se_norm": [0.6888889074325562, 0.7777777910232544], "gt_se_raw": [77.0, 80.0], "gt_se_norm": [0.7111110687255859, 0.7777777910232544], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001468.pt"}
{"idx": 1470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TmcuBRa63qg.mp4", "duration": 161.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the butterfly lands on Pooh's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the butterfly lands on Pooh's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.8136646151542664, -0.7639751434326172], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.8012422323226929, -0.7639751434326172], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001470.pt"}
{"idx": 1472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TmcuBRa63qg.mp4", "duration": 161.0, "prompt_text": "Please find the visual event described by the sentence 'At what time does Pooh take the honey pot from the cupboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does Pooh take the honey pot from the cupboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 148.0], "base_se_norm": [0.8012422323226929, 0.8385093212127686], "gt_se_raw": [145.0, 148.0], "gt_se_norm": [0.8012422323226929, 0.8385093212127686], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001472.pt"}
{"idx": 1474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AfW1MDjZB30.mp4", "duration": 47.213833333333326, "prompt_text": "Please find the visual event described by the sentence 'When does the eerie, warbling music play throughout the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the eerie, warbling music play throughout the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 46.0], "base_se_norm": [-1.0, 0.9485814571380615], "gt_se_raw": [0.0, 26.0], "gt_se_norm": [-1.0, 0.1013721227645874], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001474.pt"}
{"idx": 1476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AfW1MDjZB30.mp4", "duration": 47.213833333333326, "prompt_text": "Please find the visual event described by the sentence 'At what point does a small, rectangular object lie motionless on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a small, rectangular object lie motionless on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 25.0], "base_se_norm": [-0.23751157522201538, 0.05901169776916504], "gt_se_raw": [17.0, 26.0], "gt_se_norm": [-0.2798720598220825, 0.1013721227645874], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001476.pt"}
{"idx": 1478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AfW1MDjZB30.mp4", "duration": 47.213833333333326, "prompt_text": "Please find the visual event described by the sentence 'When can we see the question 'WHY?' appear at the bottom of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the question 'WHY?' appear at the bottom of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 46.0], "base_se_norm": [0.18609309196472168, 0.9485814571380615], "gt_se_raw": [28.0, 47.0], "gt_se_norm": [0.18609309196472168, 0.9909418821334839], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001478.pt"}
{"idx": 1480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0QswzjY6MDU.mp4", "duration": 35.035000000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a tour of the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a tour of the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.7145711183547974], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.7145711183547974], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001480.pt"}
{"idx": 1482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0QswzjY6MDU.mp4", "duration": 35.035000000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point is the person holding the bag of cauliflower pretzels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the person holding the bag of cauliflower pretzels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [0.5413157939910889, 0.7125731706619263], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.5984015464782715, 0.7125731706619263], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001482.pt"}
{"idx": 1484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JK5-2X_GqP8.mp4", "duration": 158.4583, "prompt_text": "Please find the visual event described by the sentence 'When does the boy pour vinegar into the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy pour vinegar into the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 53.0], "base_se_norm": [-0.5330001711845398, -0.3310542702674866], "gt_se_raw": [38.0, 53.0], "gt_se_norm": [-0.5203785300254822, -0.3310542702674866], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001484.pt"}
{"idx": 1486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JK5-2X_GqP8.mp4", "duration": 158.4583, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the boy add baking soda to the balloon using a funnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the boy add baking soda to the balloon using a funnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 85.0], "base_se_norm": [-0.15435165166854858, 0.0728374719619751], "gt_se_raw": [67.0, 85.0], "gt_se_norm": [-0.15435165166854858, 0.0728374719619751], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001486.pt"}
{"idx": 1488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JK5-2X_GqP8.mp4", "duration": 158.4583, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the balloon starts to inflate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the balloon starts to inflate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 108.0], "base_se_norm": [0.300026535987854, 0.36313462257385254], "gt_se_raw": [101.0, 108.0], "gt_se_norm": [0.2747833728790283, 0.36313462257385254], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001488.pt"}
{"idx": 1490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RBAmLm_jYyY.mp4", "duration": 107.32, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a person is digging in the dirt with their bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a person is digging in the dirt with their bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.7390980124473572, -0.6831904649734497], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.7577338814735413, -0.6831904649734497], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001490.pt"}
{"idx": 1492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RBAmLm_jYyY.mp4", "duration": 107.32, "prompt_text": "Please find the visual event described by the sentence 'When does the woman raise her fist in determination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman raise her fist in determination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 78.0], "base_se_norm": [0.13678717613220215, 0.45359671115875244], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [0.0808795690536499, 0.15542304515838623], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001492.pt"}
{"idx": 1494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mc5T4c_mj1U.mp4", "duration": 84.55113333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman's feet walking on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman's feet walking on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 20.0], "base_se_norm": [-0.7398024201393127, -0.5269135236740112], "gt_se_raw": [11.0, 20.0], "gt_se_norm": [-0.7398024201393127, -0.5269135236740112], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001494.pt"}
{"idx": 1496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mc5T4c_mj1U.mp4", "duration": 84.55113333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do the man and woman smile at each other by candlelight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the man and woman smile at each other by candlelight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 47.0], "base_se_norm": [0.0644446611404419, 0.11175334453582764], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.04079031944274902, 0.11175334453582764], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001496.pt"}
{"idx": 1498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mc5T4c_mj1U.mp4", "duration": 84.55113333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman looks up at the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman looks up at the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 67.0], "base_se_norm": [0.41925954818725586, 0.5848398208618164], "gt_se_raw": [60.0, 67.0], "gt_se_norm": [0.41925954818725586, 0.5848398208618164], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001498.pt"}
{"idx": 1500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZqD0iHDKmzg.mp4", "duration": 196.02916666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where Jerry hits Tom with a baseball bat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where Jerry hits Tom with a baseball bat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 83.0], "base_se_norm": [-0.18379491567611694, -0.15318721532821655], "gt_se_raw": [81.0, 83.0], "gt_se_norm": [-0.17359232902526855, -0.15318721532821655], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001500.pt"}
{"idx": 1502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZqD0iHDKmzg.mp4", "duration": 196.02916666666664, "prompt_text": "Please find the visual event described by the sentence 'When does Jerry save the fish by pulling it out of Tom's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jerry save the fish by pulling it out of Tom's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 146.0], "base_se_norm": [0.43856143951416016, 0.4895743131637573], "gt_se_raw": [142.0, 146.0], "gt_se_norm": [0.4487640857696533, 0.4895743131637573], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001502.pt"}
{"idx": 1504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eSLe4HuKuK0.mp4", "duration": 119.11899999999999, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the blue dress use her ice powers to fight the pink fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the blue dress use her ice powers to fight the pink fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 84.0], "base_se_norm": [0.3767744302749634, 0.41035425662994385], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [0.3767744302749634, 0.41035425662994385], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001504.pt"}
{"idx": 1506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eSLe4HuKuK0.mp4", "duration": 119.11899999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman ties her hair back into a ponytail.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman ties her hair back into a ponytail.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.6138315200805664, -0.5634617805480957], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [-0.6138315200805664, -0.5634617805480957], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001506.pt"}
{"idx": 1508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eSLe4HuKuK0.mp4", "duration": 119.11899999999999, "prompt_text": "Please find the visual event described by the sentence 'When does the woman on the balcony look surprised by the floating glowing crystals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman on the balcony look surprised by the floating glowing crystals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 71.0], "base_se_norm": [0.15850532054901123, 0.1920851469039917], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [0.141715407371521, 0.17529523372650146], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001508.pt"}
{"idx": 1510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l-RVtiMtfEs.mp4", "duration": 392.4921, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the NNELS logo with red hexagons is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the NNELS logo with red hexagons is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.8828001618385315, -0.8573219776153564], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.8828001618385315, -0.8573219776153564], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001510.pt"}
{"idx": 1512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l-RVtiMtfEs.mp4", "duration": 392.4921, "prompt_text": "Please find the visual event described by the sentence 'At what point is a hand-drawn map of Canada shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a hand-drawn map of Canada shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 70.0], "base_se_norm": [-0.6789743900299072, -0.6433049440383911], "gt_se_raw": [64.0, 70.0], "gt_se_norm": [-0.6738787889480591, -0.6433049440383911], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001512.pt"}
{"idx": 1514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hEteiEVRBng.mp4", "duration": 251.64, "prompt_text": "Please find the visual event described by the sentence 'When is the man sitting on a park bench and taking out a sandwich?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man sitting on a park bench and taking out a sandwich?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 13.0], "base_se_norm": [-0.9602606892585754, -0.8966777920722961], "gt_se_raw": [5.0, 11.0], "gt_se_norm": [-0.9602606892585754, -0.912573516368866], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001514.pt"}
{"idx": 1516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hEteiEVRBng.mp4", "duration": 251.64, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a masked robber with a shotgun is yelling at a store cashier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a masked robber with a shotgun is yelling at a store cashier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 97.0], "base_se_norm": [-0.32443171739578247, -0.22905737161636353], "gt_se_raw": [85.0, 93.0], "gt_se_norm": [-0.32443171739578247, -0.2608488202095032], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001516.pt"}
{"idx": 1518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hEteiEVRBng.mp4", "duration": 251.64, "prompt_text": "Please find the visual event described by the sentence 'When do several men in suits sit around a table with a wooden box on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 188.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do several men in suits sit around a table with a wooden box on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 188.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [188.0, 198.0], "base_se_norm": [0.4941980838775635, 0.5736767053604126], "gt_se_raw": [188.0, 191.0], "gt_se_norm": [0.4941980838775635, 0.5180416107177734], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001518.pt"}
{"idx": 1520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2OukFKrEWRs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator mention that people with disabilities are employed at nearly half the rate of others?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator mention that people with disabilities are employed at nearly half the rate of others?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9031662940979004, -0.6126651763916016], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.2253304123878479, 0.355671763420105], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001520.pt"}
{"idx": 1522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mD4mJF7LEJU.mp4", "duration": 30.063366666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the boy walking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy walking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9334738254547119, -0.5343169569969177], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9334738254547119, -0.5343169569969177], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001522.pt"}
{"idx": 1524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mD4mJF7LEJU.mp4", "duration": 30.063366666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time does the camera zoom into the Home Depot store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the camera zoom into the Home Depot store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.06441843509674072, 0.19747066497802734], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.0021077394485473633, 0.1309446096420288], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001524.pt"}
{"idx": 1526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mD4mJF7LEJU.mp4", "duration": 30.063366666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the 'BLUE STEEL TOOL BOX' product logo shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'BLUE STEEL TOOL BOX' product logo shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5301014184951782, 0.8627321720123291], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.5301014184951782, 0.8627321720123291], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001526.pt"}
{"idx": 1528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sIsS73PeaXI.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the reindeer is flying through the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the reindeer is flying through the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 88.0], "base_se_norm": [0.04304862022399902, 0.24038219451904297], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.8590474724769592, -0.8449522256851196], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001528.pt"}
{"idx": 1530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sIsS73PeaXI.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'At what point do two girls climb a steep ice wall using pickaxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two girls climb a steep ice wall using pickaxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 108.0], "base_se_norm": [0.3108583688735962, 0.5222872495651245], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.36723947525024414, 0.39542996883392334], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001530.pt"}
{"idx": 1532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epwyvihMOmE.mp4", "duration": 122.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the sun with a woman's face in it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the sun with a woman's face in it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 122.0], "base_se_norm": [-1.0, 0.987510085105896], "gt_se_raw": [0.0, 122.0], "gt_se_norm": [-1.0, 0.987510085105896], "start_token_span": [68, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001532.pt"}
{"idx": 1534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epwyvihMOmE.mp4", "duration": 122.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the song 'Rise and Shine' is audible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the song 'Rise and Shine' is audible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 122.0], "base_se_norm": [-1.0, 0.987510085105896], "gt_se_raw": [50.0, 60.0], "gt_se_norm": [-0.18544667959213257, -0.022535979747772217], "start_token_span": [69, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001534.pt"}
{"idx": 1536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epwyvihMOmE.mp4", "duration": 122.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman's face appear as the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman's face appear as the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 122.0], "base_se_norm": [-1.0, 0.987510085105896], "gt_se_raw": [105.0, 115.0], "gt_se_norm": [0.710561990737915, 0.8734726905822754], "start_token_span": [67, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001536.pt"}
{"idx": 1538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EUCsYSC80Xw.mp4", "duration": 98.5985985985986, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man exercising with a resistance band while sitting down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man exercising with a resistance band while sitting down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 39.0], "base_se_norm": [-0.31033504009246826, -0.20891374349594116], "gt_se_raw": [35.0, 39.0], "gt_se_norm": [-0.29005080461502075, -0.20891374349594116], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001538.pt"}
{"idx": 1540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EUCsYSC80Xw.mp4", "duration": 98.5985985985986, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a female student with a tattoo on her arm is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a female student with a tattoo on her arm is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 77.0], "base_se_norm": [0.156203031539917, 0.5618882179260254], "gt_se_raw": [56.0, 64.0], "gt_se_norm": [0.1359187364578247, 0.2981928586959839], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001540.pt"}
{"idx": 1542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8BbZTCdB2X4.mp4", "duration": 192.1, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of a kangaroo's mouth as it is chewing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of a kangaroo's mouth as it is chewing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 65.0], "base_se_norm": [-0.4169703722000122, -0.3232691287994385], "gt_se_raw": [56.0, 64.0], "gt_se_norm": [-0.4169703722000122, -0.3336803913116455], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001542.pt"}
{"idx": 1544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8BbZTCdB2X4.mp4", "duration": 192.1, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip of the newborn, hairless kangaroo crawling through the fur.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip of the newborn, hairless kangaroo crawling through the fur.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 102.0], "base_se_norm": [-0.10463303327560425, 0.061946868896484375], "gt_se_raw": [91.0, 100.0], "gt_se_norm": [-0.05257683992385864, 0.04112434387207031], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001544.pt"}
{"idx": 1546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8BbZTCdB2X4.mp4", "duration": 192.1, "prompt_text": "Please find the visual event described by the sentence 'Where can I find a shot of a large group of kangaroos grazing on a hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find a shot of a large group of kangaroos grazing on a hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 49.0], "base_se_norm": [-0.5627277493476868, -0.4898490309715271], "gt_se_raw": [42.0, 49.0], "gt_se_norm": [-0.5627277493476868, -0.4898490309715271], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001546.pt"}
{"idx": 1548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UbEEdPKBvSU.mp4", "duration": 196.125, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman slaps the man's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman slaps the man's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 150.0], "base_se_norm": [0.46845126152038574, 0.5296367406845093], "gt_se_raw": [79.0, 80.0], "gt_se_norm": [-0.19439131021499634, -0.18419373035430908], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001548.pt"}
{"idx": 1550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UbEEdPKBvSU.mp4", "duration": 196.125, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman grabs the man by his coat and pulls him close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman grabs the man by his coat and pulls him close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 150.0], "base_se_norm": [0.4582536220550537, 0.5296367406845093], "gt_se_raw": [144.0, 146.0], "gt_se_norm": [0.46845126152038574, 0.48884642124176025], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001550.pt"}
{"idx": 1552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tfIqUsJ_NA.mp4", "duration": 173.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man sings 'I can show you the world'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man sings 'I can show you the world'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 64.0], "base_se_norm": [-0.4336351752281189, -0.26025819778442383], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.884415328502655, -0.8497399091720581], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001552.pt"}
{"idx": 1554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tfIqUsJ_NA.mp4", "duration": 173.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman begins her solo part with 'A whole new world'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman begins her solo part with 'A whole new world'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 48.0], "base_se_norm": [-0.6301290988922119, -0.44519364833831787], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [-0.294933557510376, -0.26025819778442383], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001554.pt"}
{"idx": 1556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tfIqUsJ_NA.mp4", "duration": 173.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman blow a kiss to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 156.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman blow a kiss to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 156.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [156.0, 158.0], "base_se_norm": [0.8031207323074341, 0.826237678527832], "gt_se_raw": [157.0, 158.0], "gt_se_norm": [0.8146791458129883, 0.826237678527832], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001556.pt"}
{"idx": 1558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GfBVrL6WtjI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where a person is editing a video on an Apple computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where a person is editing a video on an Apple computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5409404635429382, -0.08188092708587646], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001558.pt"}
{"idx": 1560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GfBVrL6WtjI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does a man in a black t-shirt smile at the camera with his arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man in a black t-shirt smile at the camera with his arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3771786689758301, 0.8362381458282471], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001560.pt"}
{"idx": 1562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ye2sScA_bqI.mp4", "duration": 137.37056666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is the train being discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is the train being discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 18.0], "base_se_norm": [-0.8835267424583435, -0.7379351258277893], "gt_se_raw": [9.0, 18.0], "gt_se_norm": [-0.8689675331115723, -0.7379351258277893], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001562.pt"}
{"idx": 1564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ye2sScA_bqI.mp4", "duration": 137.37056666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man is using the accessibility ramp to board a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man is using the accessibility ramp to board a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 50.0], "base_se_norm": [-0.3739560842514038, -0.2720419764518738], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [-0.35939693450927734, -0.2574828267097473], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001564.pt"}
{"idx": 1566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ye2sScA_bqI.mp4", "duration": 137.37056666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a wheelchair is seen inside the bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a wheelchair is seen inside the bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 114.0], "base_se_norm": [0.2957652807235718, 0.6597442626953125], "gt_se_raw": [90.0, 102.0], "gt_se_norm": [0.31032443046569824, 0.48503434658050537], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001566.pt"}
{"idx": 1568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgWgJa1NtAY.mp4", "duration": 128.5284, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a boy is holding up and presenting the finished skateboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a boy is holding up and presenting the finished skateboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 76.0], "base_se_norm": [0.05813193321228027, 0.18261802196502686], "gt_se_raw": [69.0, 76.0], "gt_se_norm": [0.0736926794052124, 0.18261802196502686], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001568.pt"}
{"idx": 1570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgWgJa1NtAY.mp4", "duration": 128.5284, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a boy is riding the skateboard while lying on his back and then tumbles over.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a boy is riding the skateboard while lying on his back and then tumbles over.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 121.0], "base_se_norm": [0.7583663463592529, 0.8828524351119995], "gt_se_raw": [117.0, 121.0], "gt_se_norm": [0.8206093311309814, 0.8828524351119995], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001570.pt"}
{"idx": 1572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5nmhHL3sVIk.mp4", "duration": 34.7, "prompt_text": "Please find the visual event described by the sentence 'When are two children visible sitting inside the toy car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two children visible sitting inside the toy car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.15273773670196533], "gt_se_raw": [0.0, 34.0], "gt_se_norm": [-1.0, 0.9596540927886963], "start_token_span": [71, 74], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001572.pt"}
{"idx": 1574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5nmhHL3sVIk.mp4", "duration": 34.7, "prompt_text": "Please find the visual event described by the sentence 'At what time can we see a stuffed dog on the roof of the vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time can we see a stuffed dog on the roof of the vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.15273773670196533], "gt_se_raw": [2.0, 20.0], "gt_se_norm": [-0.8847262263298035, 0.15273773670196533], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001574.pt"}
{"idx": 1576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5nmhHL3sVIk.mp4", "duration": 34.7, "prompt_text": "Please find the visual event described by the sentence 'When does the child in the driver's seat lean forward and look out the front window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child in the driver's seat lean forward and look out the front window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 34.0], "base_se_norm": [0.2680114507675171, 0.9596540927886963], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [0.7291065454483032, 0.9020172357559204], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001576.pt"}
{"idx": 1578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Uo0KjdDJr1c.mp4", "duration": 170.83733333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman lists the social media apps she is proficient in.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman lists the social media apps she is proficient in.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 36.0], "base_se_norm": [-0.800980269908905, -0.578546404838562], "gt_se_raw": [18.0, 24.0], "gt_se_norm": [-0.789273202419281, -0.7190309762954712], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001578.pt"}
{"idx": 1580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Uo0KjdDJr1c.mp4", "duration": 170.83733333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is seen walking away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is seen walking away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 128.0], "base_se_norm": [0.4399663209915161, 0.4985015392303467], "gt_se_raw": [123.0, 127.0], "gt_se_norm": [0.4399663209915161, 0.48679447174072266], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001580.pt"}
{"idx": 1582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GyTxtJ4gVLE.mp4", "duration": 110.54387721054388, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker introduce herself and state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker introduce herself and state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9638152718544006], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9819076061248779, -0.9457229375839233], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001582.pt"}
{"idx": 1584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GyTxtJ4gVLE.mp4", "duration": 110.54387721054388, "prompt_text": "Please find the visual event described by the sentence 'At what point are the body parts affected by tremors, such as the hand, legs, and jaw, shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the body parts affected by tremors, such as the hand, legs, and jaw, shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 45.0], "base_se_norm": [-0.24012070894241333, -0.18584364652633667], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [-0.22202837467193604, -0.1677512526512146], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001584.pt"}
{"idx": 1586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GyTxtJ4gVLE.mp4", "duration": 110.54387721054388, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the Liftware Steady utensil is being used to pick up a cherry tomato from a salad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the Liftware Steady utensil is being used to pick up a cherry tomato from a salad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 81.0], "base_se_norm": [0.32074248790740967, 0.46548140048980713], "gt_se_raw": [73.0, 81.0], "gt_se_norm": [0.32074248790740967, 0.46548140048980713], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001586.pt"}
{"idx": 1588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fnd-2jetT1w.mp4", "duration": 259.92633333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a cartoon character in a yellow shirt is standing next to a blackboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a cartoon character in a yellow shirt is standing next to a blackboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [-0.6845259666442871, -0.6537480354309082], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [-0.6845259666442871, -0.6537480354309082], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001588.pt"}
{"idx": 1590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fnd-2jetT1w.mp4", "duration": 259.92633333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a person is mixing cornstarch and water in a pink bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 181.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a person is mixing cornstarch and water in a pink bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 181.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 181.0], "base_se_norm": [0.33114635944366455, 0.3927023410797119], "gt_se_raw": [173.0, 179.0], "gt_se_norm": [0.33114635944366455, 0.37731337547302246], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001590.pt"}
{"idx": 1592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PwirCe0230M.mp4", "duration": 33.70033333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the camera pan to show the bed on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera pan to show the bed on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.34718745946884155, -0.05045449733734131], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.34718745946884155, -0.05045449733734131], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001592.pt"}
{"idx": 1594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PwirCe0230M.mp4", "duration": 33.70033333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman turn her head towards the camera and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman turn her head towards the camera and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [0.2462785243988037, 0.4243183135986328], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.3056250810623169, 0.4243183135986328], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001594.pt"}
{"idx": 1596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PwirCe0230M.mp4", "duration": 33.70033333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the printer actively printing the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the printer actively printing the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [0.6617046594619751, 0.9584376811981201], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [0.8397444486618042, 0.9584376811981201], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001596.pt"}
{"idx": 1598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6d3zmkbef28.mp4", "duration": 39.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the period where the giraffe is standing still in the middle of the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period where the giraffe is standing still in the middle of the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 29.0], "base_se_norm": [-0.8983911871910095, 0.47332775592803955], "gt_se_raw": [2.0, 29.0], "gt_se_norm": [-0.8983911871910095, 0.47332775592803955], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001598.pt"}
{"idx": 1600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6d3zmkbef28.mp4", "duration": 39.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where a baby giraffe is seen walking in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where a baby giraffe is seen walking in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 38.0], "base_se_norm": [0.6257410049438477, 0.9305673837661743], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [0.6765453815460205, 0.8289586305618286], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001600.pt"}
{"idx": 1602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R9Puz5RFl5o.mp4", "duration": 82.816, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the teacher helps the boy draw patterns inside the object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the teacher helps the boy draw patterns inside the object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 45.0], "base_se_norm": [-0.5170015692710876, 0.0867464542388916], "gt_se_raw": [27.0, 33.0], "gt_se_norm": [-0.34795212745666504, -0.2030525803565979], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001602.pt"}
{"idx": 1604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R9Puz5RFl5o.mp4", "duration": 82.816, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the boy's hand to draw a long straight line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the boy's hand to draw a long straight line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [-0.009853184223175049, 0.0867464542388916], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [0.014296770095825195, 0.06259655952453613], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001604.pt"}
{"idx": 1606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R9Puz5RFl5o.mp4", "duration": 82.816, "prompt_text": "Please find the visual event described by the sentence 'When does the woman lift the completed drawing off the clipboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman lift the completed drawing off the clipboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.5214451551437378, 0.6421947479248047], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [0.5214451551437378, 0.5697449445724487], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001606.pt"}
{"idx": 1608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/llsWcPkQZJ4.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Show the rooster crowing loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the rooster crowing loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 53.0], "base_se_norm": [-0.28404927253723145, -0.1175491213798523], "gt_se_raw": [45.0, 53.0], "gt_se_norm": [-0.25074928998947144, -0.1175491213798523], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001608.pt"}
{"idx": 1610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/llsWcPkQZJ4.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'When can we see a baby chicken hatching from an egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a baby chicken hatching from an egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 92.0], "base_se_norm": [0.46520137786865234, 0.5318014621734619], "gt_se_raw": [88.0, 92.0], "gt_se_norm": [0.46520137786865234, 0.5318014621734619], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001610.pt"}
{"idx": 1612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nxzo7WLQ50s.mp4", "duration": 13.446766666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the child struggling to open the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the child struggling to open the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.40506142377853394, 0.04114246368408203], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.2563267946243286], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001612.pt"}
{"idx": 1614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nxzo7WLQ50s.mp4", "duration": 13.446766666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child take a pink piece of clothing out of the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child take a pink piece of clothing out of the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [0.18987715244293213, 0.6360809803009033], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [0.18987715244293213, 0.6360809803009033], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001614.pt"}
{"idx": 1616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nxzo7WLQ50s.mp4", "duration": 13.446766666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the child in the orange shirt is interacting with the cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the child in the orange shirt is interacting with the cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, 0.933550238609314], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, 0.933550238609314], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001616.pt"}
{"idx": 1618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JMKHbmwltWQ.mp4", "duration": 222.6224, "prompt_text": "Please find the visual event described by the sentence 'At what point is the blindfolded person being helped to sit down in a chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the blindfolded person being helped to sit down in a chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.883210301399231, -0.8382912278175354], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.8742265105247498, -0.8382912278175354], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001618.pt"}
{"idx": 1620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JMKHbmwltWQ.mp4", "duration": 222.6224, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the two individuals are interacting with a clear plastic bag and its contents.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the two individuals are interacting with a clear plastic bag and its contents.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 78.0], "base_se_norm": [-0.47893834114074707, -0.2992619276046753], "gt_se_raw": [58.0, 81.0], "gt_se_norm": [-0.47893834114074707, -0.27231043577194214], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001620.pt"}
{"idx": 1622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CY0629SKvt8.mp4", "duration": 54.054, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in purple place the colorful plate on the boy's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in purple place the colorful plate on the boy's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.777999758720398], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9259999394416809], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001622.pt"}
{"idx": 1624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CY0629SKvt8.mp4", "duration": 54.054, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the boy is pulling the beads with both hands, causing them to make a rattling sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the boy is pulling the beads with both hands, causing them to make a rattling sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 30.0], "base_se_norm": [-0.5189995169639587, 0.1100010871887207], "gt_se_raw": [21.0, 30.0], "gt_se_norm": [-0.2229992151260376, 0.1100010871887207], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001624.pt"}
{"idx": 1626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CY0629SKvt8.mp4", "duration": 54.054, "prompt_text": "Please find the visual event described by the sentence 'When does the woman pick up the beads from the floor and put them back on the plate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman pick up the beads from the floor and put them back on the plate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 53.0], "base_se_norm": [0.665001630783081, 0.9610018730163574], "gt_se_raw": [47.0, 53.0], "gt_se_norm": [0.7390017509460449, 0.9610018730163574], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001626.pt"}
{"idx": 1628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFY-vIeN5XA.mp4", "duration": 57.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a girl shows the number 16 painted on her palms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a girl shows the number 16 painted on her palms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [-0.2704111337661743, -0.16618412733078003], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.23566877841949463, -0.16618412733078003], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001628.pt"}
{"idx": 1630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFY-vIeN5XA.mp4", "duration": 57.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the number 16 is painted on the soles of someone's feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the number 16 is painted on the soles of someone's feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.23566877841949463, -0.13144177198410034], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [0.04226982593536377, 0.11175453662872314], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001630.pt"}
{"idx": 1632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dv0VfX9U9Fg.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a wheelchair having a business meeting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a wheelchair having a business meeting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 11.0], "base_se_norm": [-0.7742370367050171, -0.3791518807411194], "gt_se_raw": [3.0, 11.0], "gt_se_norm": [-0.8306778073310852, -0.3791518807411194], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001632.pt"}
{"idx": 1634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dv0VfX9U9Fg.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'At what point is an animated map of Minnesota being filled with icons of people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an animated map of Minnesota being filled with icons of people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.15338891744613647, 0.07237398624420166], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [-0.2098296880722046, 0.07237398624420166], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001634.pt"}
{"idx": 1636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dv0VfX9U9Fg.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the CareerForceMN.com website is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the CareerForceMN.com website is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [0.3545776605606079, 0.6367813348770142], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [0.29813694953918457, 0.6367813348770142], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001636.pt"}
{"idx": 1638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6D4B81EEr30.mp4", "duration": 33.166466666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when the boy pauses and leans his head on his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when the boy pauses and leans his head on his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.38694310188293457, 0.5678486824035645], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.4472450017929077, 0.5678486824035645], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001638.pt"}
{"idx": 1640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6D4B81EEr30.mp4", "duration": 33.166466666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time is the woman's scarf most clearly visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the woman's scarf most clearly visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 32.0], "base_se_norm": [-1.0, 0.9296599626541138], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.758792519569397], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001640.pt"}
{"idx": 1642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DF8_5mGdcq4.mp4", "duration": 79.954875, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where abstract and eerie music plays over a black screen before any person appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where abstract and eerie music plays over a black screen before any person appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 47.0], "base_se_norm": [-1.0, 0.17566323280334473], "gt_se_raw": [0.0, 16.0], "gt_se_norm": [-1.0, -0.5997742414474487], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001642.pt"}
{"idx": 1644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DF8_5mGdcq4.mp4", "duration": 79.954875, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman open her glowing eyes and stare at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman open her glowing eyes and stare at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 57.0], "base_se_norm": [0.22569143772125244, 0.4258042573928833], "gt_se_raw": [50.0, 57.0], "gt_se_norm": [0.2507054805755615, 0.4258042573928833], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001644.pt"}
{"idx": 1646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DF8_5mGdcq4.mp4", "duration": 79.954875, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'SUBSCRIBE TO DAD' show up on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'SUBSCRIBE TO DAD' show up on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 79.0], "base_se_norm": [0.5008466243743896, 0.9761147499084473], "gt_se_raw": [60.0, 79.0], "gt_se_norm": [0.5008466243743896, 0.9761147499084473], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001646.pt"}
{"idx": 1648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dRaKMNYG3-I.mp4", "duration": 171.62979166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman peeks out of the door and sees a blurred figure in the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman peeks out of the door and sees a blurred figure in the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 56.0], "base_se_norm": [-0.510574460029602, -0.34743261337280273], "gt_se_raw": [41.0, 56.0], "gt_se_norm": [-0.5222274661064148, -0.34743261337280273], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001648.pt"}
{"idx": 1650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dRaKMNYG3-I.mp4", "duration": 171.62979166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman puts on her glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman puts on her glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 71.0], "base_se_norm": [-0.21924978494644165, -0.17263782024383545], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [-0.2075967788696289, -0.17263782024383545], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001650.pt"}
{"idx": 1652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZtCZGwLH5_o.mp4", "duration": 118.48503333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy is excited with his mouth wide open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy is excited with his mouth wide open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 48.0], "base_se_norm": [-0.44296759366989136, -0.18977105617523193], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.4598473906517029, -0.30792945623397827], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001652.pt"}
{"idx": 1654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZtCZGwLH5_o.mp4", "duration": 118.48503333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when someone holds the boy's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when someone holds the boy's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 27.0], "base_se_norm": [-0.7805629968643188, -0.5442461967468262], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.7805629968643188, -0.7299237251281738], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001654.pt"}
{"idx": 1656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZtCZGwLH5_o.mp4", "duration": 118.48503333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy successfully stand up from the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy successfully stand up from the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 108.0], "base_se_norm": [0.7554960250854492, 0.8230150938034058], "gt_se_raw": [104.0, 107.0], "gt_se_norm": [0.7554960250854492, 0.8061352968215942], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001656.pt"}
{"idx": 1658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2I943-gP904.mp4", "duration": 60.685624999999995, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman in a grey sweatsuit is doing sit-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman in a grey sweatsuit is doing sit-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 27.0], "base_se_norm": [-0.3408653140068054, -0.11016815900802612], "gt_se_raw": [20.0, 27.0], "gt_se_norm": [-0.3408653140068054, -0.11016815900802612], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001658.pt"}
{"idx": 1660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2I943-gP904.mp4", "duration": 60.685624999999995, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an elderly man is happily talking on his phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an elderly man is happily talking on his phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 45.0], "base_se_norm": [0.38418281078338623, 0.4830530881881714], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [0.3512260913848877, 0.4830530881881714], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001660.pt"}
{"idx": 1662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFxrHOo3NJ0.mp4", "duration": 224.09053333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the black wetsuit speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the black wetsuit speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.919675350189209], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, -0.919675350189209], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001662.pt"}
{"idx": 1664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFxrHOo3NJ0.mp4", "duration": 224.09053333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator identify a fish with only one eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 178.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator identify a fish with only one eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 178.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [178.0, 182.0], "base_se_norm": [0.5886436700820923, 0.6243435144424438], "gt_se_raw": [82.0, 88.0], "gt_se_norm": [-0.2681528925895691, -0.21460312604904175], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001664.pt"}
{"idx": 1666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFxrHOo3NJ0.mp4", "duration": 224.09053333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the diver waving goodbye at the end of the video.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 220.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the diver waving goodbye at the end of the video.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 220.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [220.0, 222.0], "base_se_norm": [0.9634921550750732, 0.981342077255249], "gt_se_raw": [220.0, 222.0], "gt_se_norm": [0.9634921550750732, 0.981342077255249], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001666.pt"}
{"idx": 1668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5TiaQeuNO0.mp4", "duration": 178.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the green tunic throws the dagger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the green tunic throws the dagger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [-0.6974225044250488, -0.6750093698501587], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [-0.686215877532959, -0.6638027429580688], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001668.pt"}
{"idx": 1670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5TiaQeuNO0.mp4", "duration": 178.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man in green says 'Stop saying that!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man in green says 'Stop saying that!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 61.0], "base_se_norm": [-0.37243181467056274, -0.3163989186286926], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [0.5240942239761353, 0.5465073585510254], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001670.pt"}
{"idx": 1672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WkxE_Fs_mHI.mp4", "duration": 157.07358333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in the red dress and the man dancing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in the red dress and the man dancing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 68.0], "base_se_norm": [-0.2360268235206604, -0.1341637372970581], "gt_se_raw": [61.0, 67.0], "gt_se_norm": [-0.2232939600944519, -0.1468966007232666], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001672.pt"}
{"idx": 1674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WkxE_Fs_mHI.mp4", "duration": 157.07358333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman is holding a fan to her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman is holding a fan to her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 85.0], "base_se_norm": [-0.006834864616394043, 0.08229529857635498], "gt_se_raw": [78.0, 85.0], "gt_se_norm": [-0.006834864616394043, 0.08229529857635498], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001674.pt"}
{"idx": 1676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WkxE_Fs_mHI.mp4", "duration": 157.07358333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman feeds the man a grape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman feeds the man a grape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [-0.3124241232872009, -0.2360268235206604], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.3124241232872009, -0.26149260997772217], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001676.pt"}
{"idx": 1678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vz3zueYyac8.mp4", "duration": 106.70660000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see an underwater view of a large wave curling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see an underwater view of a large wave curling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.6626262664794922, -0.5876543521881104], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.6626262664794922, -0.6063973307609558], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001678.pt"}
{"idx": 1680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vz3zueYyac8.mp4", "duration": 106.70660000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the whale shark open its mouth wide underwater to feed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the whale shark open its mouth wide underwater to feed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 68.0], "base_se_norm": [-0.15656572580337524, 0.27452290058135986], "gt_se_raw": [49.0, 58.0], "gt_se_norm": [-0.08159381151199341, 0.08709311485290527], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001680.pt"}
{"idx": 1682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FXkwv_zWL84.mp4", "duration": 197.3972, "prompt_text": "Please find the visual event described by the sentence 'When does the person start pouring cornstarch into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person start pouring cornstarch into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 63.0], "base_se_norm": [-0.4934071898460388, -0.3616930842399597], "gt_se_raw": [51.0, 63.0], "gt_se_norm": [-0.4832753539085388, -0.3616930842399597], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001682.pt"}
{"idx": 1684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FXkwv_zWL84.mp4", "duration": 197.3972, "prompt_text": "Please find the visual event described by the sentence 'At what point is the blue and red food coloring added to the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the blue and red food coloring added to the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 81.0], "base_se_norm": [-0.2806382179260254, -0.17931967973709106], "gt_se_raw": [71.0, 81.0], "gt_se_norm": [-0.2806382179260254, -0.17931967973709106], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001684.pt"}
{"idx": 1686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FXkwv_zWL84.mp4", "duration": 197.3972, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is playing with the finished oobleck with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is playing with the finished oobleck with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 169.0], "base_se_norm": [0.39819610118865967, 0.7122836112976074], "gt_se_raw": [138.0, 149.0], "gt_se_norm": [0.39819610118865967, 0.5096465349197388], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001686.pt"}
{"idx": 1688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lrfIVnMGXK0.mp4", "duration": 180.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man lift the couch with the bear lying on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man lift the couch with the bear lying on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.7333333492279053, -0.6666666269302368], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.7222222089767456, -0.6111111044883728], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001688.pt"}
{"idx": 1690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lrfIVnMGXK0.mp4", "duration": 180.0, "prompt_text": "Please find the visual event described by the sentence 'When does the man show off his eight-pack abs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man show off his eight-pack abs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 77.0], "base_se_norm": [-0.19999998807907104, -0.14444446563720703], "gt_se_raw": [73.0, 77.0], "gt_se_norm": [-0.18888890743255615, -0.14444446563720703], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001690.pt"}
{"idx": 1692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IApvU6SMq-8.mp4", "duration": 217.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man walking down the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man walking down the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 106.0], "base_se_norm": [-0.9723587036132812, -0.02334153652191162], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.9631449580192566, -0.9170761704444885], "start_token_span": [65, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001692.pt"}
{"idx": 1694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IApvU6SMq-8.mp4", "duration": 217.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a blue food truck is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a blue food truck is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 82.0], "base_se_norm": [-0.3550368547439575, -0.2444717288017273], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [-0.34582310914993286, -0.30896806716918945], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001694.pt"}
{"idx": 1696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IApvU6SMq-8.mp4", "duration": 217.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two women, one wearing a hijab, are smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 183.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two women, one wearing a hijab, are smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 183.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [183.0, 185.0], "base_se_norm": [0.6861178874969482, 0.7045454978942871], "gt_se_raw": [184.0, 185.0], "gt_se_norm": [0.6953316926956177, 0.7045454978942871], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001696.pt"}
{"idx": 1698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggIzq8JUK44.mp4", "duration": 185.97745833333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man bites into the carrots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man bites into the carrots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 100.0], "base_se_norm": [0.0216292142868042, 0.07539916038513184], "gt_se_raw": [95.0, 98.0], "gt_se_norm": [0.0216292142868042, 0.05389118194580078], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001698.pt"}
{"idx": 1700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggIzq8JUK44.mp4", "duration": 185.97745833333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the man simulates the sound of fairy wings with a bicycle wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the man simulates the sound of fairy wings with a bicycle wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 134.0], "base_se_norm": [0.35500288009643555, 0.44103479385375977], "gt_se_raw": [126.0, 130.0], "gt_se_norm": [0.35500288009643555, 0.39801883697509766], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001700.pt"}
{"idx": 1702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jyhFQVEMhD0.mp4", "duration": 276.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player in green build a wall around the central platform?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in green build a wall around the central platform?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.8266104459762573, -0.7977122068405151], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.8193858861923218, -0.7977122068405151], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001702.pt"}
{"idx": 1704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jyhFQVEMhD0.mp4", "duration": 276.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the trapped player tries to bribe their way out with a diamond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the trapped player tries to bribe their way out with a diamond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 101.0], "base_se_norm": [-0.2992173433303833, -0.2703191041946411], "gt_se_raw": [97.0, 101.0], "gt_se_norm": [-0.2992173433303833, -0.2703191041946411], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001704.pt"}
{"idx": 1706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jyhFQVEMhD0.mp4", "duration": 276.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the silverfish spawned, leading to a player's elimination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 236.0 - 256.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the silverfish spawned, leading to a player's elimination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 236.0 - 256.0 seconds.<|im_end|>\n", "base_se_raw": [236.0, 256.0], "base_se_norm": [0.7049969434738159, 0.8494881391525269], "gt_se_raw": [234.0, 243.0], "gt_se_norm": [0.6905478239059448, 0.7555688619613647], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001706.pt"}
{"idx": 1708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wn13H7OzGGI.mp4", "duration": 77.91116666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman says 'Dad, daughter'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman says 'Dad, daughter'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.28123265504837036, -0.17855161428451538], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.7689676284790039, -0.6919568777084351], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001708.pt"}
{"idx": 1710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wn13H7OzGGI.mp4", "duration": 77.91116666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman hug the bone to her chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman hug the bone to her chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [-0.15288132429122925, -0.02453005313873291], "gt_se_raw": [33.0, 39.0], "gt_se_norm": [-0.15288132429122925, 0.0011402368545532227], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001710.pt"}
{"idx": 1712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M0nDEbrp9nM.mp4", "duration": 157.19704, "prompt_text": "Please find the visual event described by the sentence 'When can we see the blue alien being taken out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the blue alien being taken out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 20.0], "base_se_norm": [-0.860048234462738, -0.7455422878265381], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.860048234462738, -0.8218796253204346], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001712.pt"}
{"idx": 1714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M0nDEbrp9nM.mp4", "duration": 157.19704, "prompt_text": "Please find the visual event described by the sentence 'At what point does the little girl place a flower on the blue alien's chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the little girl place a flower on the blue alien's chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [-0.31296414136886597, -0.21118104457855225], "gt_se_raw": [55.0, 62.0], "gt_se_norm": [-0.30024129152297974, -0.21118104457855225], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001714.pt"}
{"idx": 1716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M0nDEbrp9nM.mp4", "duration": 157.19704, "prompt_text": "Please find the visual event described by the sentence 'When are all four characters standing together looking sad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all four characters standing together looking sad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [-0.02033776044845581, 0.0559995174407959], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.02033776044845581, 0.030553817749023438], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001716.pt"}
{"idx": 1718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nIs4S9YDPRs.mp4", "duration": 147.78096666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Spider-Man drop into the truck and confront the thief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Spider-Man drop into the truck and confront the thief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 98.0], "base_se_norm": [-0.07971906661987305, 0.32628726959228516], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [-0.29625576734542847, -0.25565510988235474], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001718.pt"}
{"idx": 1720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nIs4S9YDPRs.mp4", "duration": 147.78096666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the sand villain lift and overturn the armored truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the sand villain lift and overturn the armored truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 110.0], "base_se_norm": [0.4345555305480957, 0.4886897802352905], "gt_se_raw": [107.0, 111.0], "gt_se_norm": [0.4480891227722168, 0.5022232532501221], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001720.pt"}
{"idx": 1722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y4baNTo5q4A.mp4", "duration": 62.020291666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the woman looking up with a worried expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman looking up with a worried expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 18.0], "base_se_norm": [-0.7742674350738525, -0.4195448160171509], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.7742674350738525, -0.6775249242782593], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001722.pt"}
{"idx": 1724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y4baNTo5q4A.mp4", "duration": 62.020291666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman walks from left to right across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman walks from left to right across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [0.03192031383514404, 0.19315791130065918], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [0.03192031383514404, 0.19315791130065918], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001724.pt"}
{"idx": 1726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y4baNTo5q4A.mp4", "duration": 62.020291666666665, "prompt_text": "Please find the visual event described by the sentence 'Identify when the text 'SUBSCRIBE TO DAD' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify when the text 'SUBSCRIBE TO DAD' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 61.0], "base_se_norm": [0.41889047622680664, 0.9670981168746948], "gt_se_raw": [44.0, 62.0], "gt_se_norm": [0.41889047622680664, 0.9993456602096558], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001726.pt"}
{"idx": 1728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ppe_c4Ij0g8.mp4", "duration": 154.45445445445446, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a little girl with blonde hair gives an elderly man a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a little girl with blonde hair gives an elderly man a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 63.0], "base_se_norm": [-0.27486711740493774, -0.18422549962997437], "gt_se_raw": [55.0, 63.0], "gt_se_norm": [-0.2878159284591675, -0.18422549962997437], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001728.pt"}
{"idx": 1730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ppe_c4Ij0g8.mp4", "duration": 154.45445445445446, "prompt_text": "Please find the visual event described by the sentence 'At what time is a photo of three soldiers standing in front of a tank in the woods displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a photo of three soldiers standing in front of a tank in the woods displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [0.07475054264068604, 0.10064804553985596], "gt_se_raw": [82.0, 85.0], "gt_se_norm": [0.06180167198181152, 0.10064804553985596], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001730.pt"}
{"idx": 1732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RAqA6gtdalE.mp4", "duration": 150.96, "prompt_text": "Please find the visual event described by the sentence 'Show me a clip where someone is whispering 'Paris'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me a clip where someone is whispering 'Paris'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.8675146102905273, -0.841017484664917], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9867514371871948, -0.9470058083534241], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001732.pt"}
{"idx": 1734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RAqA6gtdalE.mp4", "duration": 150.96, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when rhythmic whispering and clicking noises can be heard together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when rhythmic whispering and clicking noises can be heard together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.8675146102905273, -0.8012718558311462], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.642289400100708, -0.5892952084541321], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001734.pt"}
{"idx": 1736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RAqA6gtdalE.mp4", "duration": 150.96, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the whispering voice repeats 'Paris' several times in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the whispering voice repeats 'Paris' several times in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [0.3248542547225952, 0.4838367700576782], "gt_se_raw": [83.0, 86.0], "gt_se_norm": [0.09962904453277588, 0.13937461376190186], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001736.pt"}
{"idx": 1738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vv2qnoUfjPU.mp4", "duration": 142.39225, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man with a beard and a black t-shirt points at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man with a beard and a black t-shirt points at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.690994381904602, -0.6488572359085083], "gt_se_raw": [23.0, 24.0], "gt_se_norm": [-0.6769486665725708, -0.6629029512405396], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001738.pt"}
{"idx": 1740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vv2qnoUfjPU.mp4", "duration": 142.39225, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a man with long hair in a ponytail is speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a man with long hair in a ponytail is speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 57.0], "base_se_norm": [-0.3117602467536926, -0.1993945837020874], "gt_se_raw": [50.0, 57.0], "gt_se_norm": [-0.2977145314216614, -0.1993945837020874], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001740.pt"}
{"idx": 1742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EpcDZbXslfw.mp4", "duration": 180.71031812826124, "prompt_text": "Please find the visual event described by the sentence 'When is the woman bitten by the snake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman bitten by the snake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 49.0], "base_se_norm": [-0.6458420157432556, -0.4576956033706665], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.7122466564178467, -0.6790443658828735], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001742.pt"}
{"idx": 1744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EpcDZbXslfw.mp4", "duration": 180.71031812826124, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the car pulls up to the hospital emergency entrance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the car pulls up to the hospital emergency entrance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.2695491313934326, -0.21421194076538086], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.2695491313934326, -0.21421194076538086], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001744.pt"}
{"idx": 1746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EpcDZbXslfw.mp4", "duration": 180.71031812826124, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man on the sofa reacts in shock to the hospital bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 168.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man on the sofa reacts in shock to the hospital bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 168.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [168.0, 172.0], "base_se_norm": [0.8593294620513916, 0.9035991430282593], "gt_se_raw": [168.0, 172.0], "gt_se_norm": [0.8593294620513916, 0.9035991430282593], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001746.pt"}
{"idx": 1748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NvgYhf2LnVI.mp4", "duration": 145.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a person is sinking in the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a person is sinking in the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [-0.13560479879379272, -0.08072257041931152], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [-0.14932537078857422, -0.09444314241409302], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001748.pt"}
{"idx": 1750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NvgYhf2LnVI.mp4", "duration": 145.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the title 'DREAM SMP WAR' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the title 'DREAM SMP WAR' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 143.0], "base_se_norm": [0.9208781719207764, 0.9620398283004761], "gt_se_raw": [140.0, 142.0], "gt_se_norm": [0.9208781719207764, 0.9483193159103394], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001750.pt"}
{"idx": 1752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rvsuYssoTw0.mp4", "duration": 158.95045833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man in a beige shirt is tackled and falls in the restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man in a beige shirt is tackled and falls in the restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.7609317898750305, -0.7106016874313354], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.7483492493629456, -0.7231841683387756], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001752.pt"}
{"idx": 1754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rvsuYssoTw0.mp4", "duration": 158.95045833333333, "prompt_text": "Please find the visual event described by the sentence 'A man in a light blue shirt falls to the floor in an office setting. Pinpoint the time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a light blue shirt falls to the floor in an office setting. Pinpoint the time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 70.0], "base_se_norm": [-0.15697002410888672, -0.11922240257263184], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.4966984987258911, -0.4715334177017212], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001754.pt"}
{"idx": 1756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rvsuYssoTw0.mp4", "duration": 158.95045833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the grey t-shirt headbutts the man in the blue 'Italia' shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the grey t-shirt headbutts the man in the blue 'Italia' shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 142.0], "base_se_norm": [0.7489726543426514, 0.7867202758789062], "gt_se_raw": [138.0, 141.0], "gt_se_norm": [0.7363901138305664, 0.7741377353668213], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001756.pt"}
{"idx": 1758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KK-Mff60ZIE.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the presenter introduces the Microsoft Adaptive Controller as a solution.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the presenter introduces the Microsoft Adaptive Controller as a solution.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 48.0], "base_se_norm": [-0.5756906270980835, -0.36353588104248047], "gt_se_raw": [23.0, 32.0], "gt_se_norm": [-0.6950275897979736, -0.5756906270980835], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001758.pt"}
{"idx": 1760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KK-Mff60ZIE.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where robotic arms are testing game controllers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where robotic arms are testing game controllers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 17.0], "base_se_norm": [-0.8541436195373535, -0.7745856046676636], "gt_se_raw": [10.0, 17.0], "gt_se_norm": [-0.8674033284187317, -0.7745856046676636], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001760.pt"}
{"idx": 1762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9newNjIco3Y.mp4", "duration": 121.6, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy touching the tablet with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy touching the tablet with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 10.0], "base_se_norm": [-0.9671052694320679, -0.8355263471603394], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.2598683834075928, -0.22697365283966064], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001762.pt"}
{"idx": 1764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9newNjIco3Y.mp4", "duration": 121.6, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man holding the boy's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man holding the boy's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.7697368264198303, -0.7039473652839661], "gt_se_raw": [93.0, 97.0], "gt_se_norm": [0.5296052694320679, 0.5953947305679321], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001764.pt"}
{"idx": 1766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9newNjIco3Y.mp4", "duration": 121.6, "prompt_text": "Please find the visual event described by the sentence 'When is the boy wearing a striped shirt and looking at the tablet with his mouth open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy wearing a striped shirt and looking at the tablet with his mouth open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 28.0], "base_se_norm": [-0.8190789222717285, -0.5394736528396606], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [-0.6381579041481018, -0.5723683834075928], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001766.pt"}
{"idx": 1768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dx0-pNkwOv8.mp4", "duration": 139.80633333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a movie poster for 'Wonder' is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a movie poster for 'Wonder' is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 20.0], "base_se_norm": [-0.8569449782371521, -0.7138899564743042], "gt_se_raw": [10.0, 21.0], "gt_se_norm": [-0.8569449782371521, -0.699584424495697], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001768.pt"}
{"idx": 1770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dx0-pNkwOv8.mp4", "duration": 139.80633333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video illustrate a disabled student being rejected for enrollment by a school?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video illustrate a disabled student being rejected for enrollment by a school?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 81.0], "base_se_norm": [0.0013852119445800781, 0.15874576568603516], "gt_se_raw": [70.0, 81.0], "gt_se_norm": [0.0013852119445800781, 0.15874576568603516], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001770.pt"}
{"idx": 1772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lGcSN2H8MNk.mp4", "duration": 27.3273, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman chopping green peppers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman chopping green peppers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.5608786344528198, -0.3413180112838745], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.4876917600631714, -0.3413180112838745], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001772.pt"}
{"idx": 1774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lGcSN2H8MNk.mp4", "duration": 27.3273, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an instructor is explaining the recipe to the students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an instructor is explaining the recipe to the students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [0.3905508518218994, 0.6832984685897827], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.46373772621154785, 0.6832984685897827], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001774.pt"}
{"idx": 1776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lGcSN2H8MNk.mp4", "duration": 27.3273, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman is cutting a green mango with a large knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman is cutting a green mango with a large knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [0.7564853429794312, 0.902859091758728], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.7564853429794312, 0.902859091758728], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001776.pt"}
{"idx": 1778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y07at1bU89Q.mp4", "duration": 155.04, "prompt_text": "Please find the visual event described by the sentence 'When does the dog successfully guide the group of men through the two white gates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog successfully guide the group of men through the two white gates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [-0.44530439376831055, -0.38080495595932007], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.4324045181274414, -0.38080495595932007], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001778.pt"}
{"idx": 1780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y07at1bU89Q.mp4", "duration": 155.04, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when three men break away from the group and run towards the stage with the dancing women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when three men break away from the group and run towards the stage with the dancing women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 108.0], "base_se_norm": [0.27708983421325684, 0.3931889533996582], "gt_se_raw": [99.0, 103.0], "gt_se_norm": [0.27708983421325684, 0.32868945598602295], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001780.pt"}
{"idx": 1782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/agqgBkpbCoY.mp4", "duration": 164.8647, "prompt_text": "Please find the visual event described by the sentence 'Show me when the video presents a 3D model of a shark with its body parts labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the video presents a 3D model of a shark with its body parts labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 37.0], "base_se_norm": [-0.6603275537490845, -0.5511471033096313], "gt_se_raw": [28.0, 37.0], "gt_se_norm": [-0.6603275537490845, -0.5511471033096313], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001782.pt"}
{"idx": 1784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/agqgBkpbCoY.mp4", "duration": 164.8647, "prompt_text": "Please find the visual event described by the sentence 'At what point does a great white shark leap out of the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a great white shark leap out of the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 75.0], "base_se_norm": [-0.17508113384246826, -0.09016305208206177], "gt_se_raw": [68.0, 75.0], "gt_se_norm": [-0.17508113384246826, -0.09016305208206177], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001784.pt"}
{"idx": 1786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/agqgBkpbCoY.mp4", "duration": 164.8647, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where a close-up shows a shark's gills?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where a close-up shows a shark's gills?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 97.0], "base_se_norm": [0.1403290033340454, 0.17672252655029297], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.1403290033340454, 0.17672252655029297], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001786.pt"}
{"idx": 1788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YFB1moyD1UY.mp4", "duration": 113.6, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the speaker clarifies that the services are available for children with either diagnosed or undiagnosed disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the speaker clarifies that the services are available for children with either diagnosed or undiagnosed disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [-0.19014084339141846, -0.0845070481300354], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.5246478915214539, -0.47183096408843994], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001788.pt"}
{"idx": 1790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YFB1moyD1UY.mp4", "duration": 113.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention providing advocacy services by going into school meetings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention providing advocacy services by going into school meetings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.17253518104553223, -0.0845070481300354], "gt_se_raw": [77.0, 84.0], "gt_se_norm": [0.35563385486602783, 0.47887325286865234], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001790.pt"}
{"idx": 1792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcNcX0hEr_s.mp4", "duration": 88.1, "prompt_text": "Please find the visual event described by the sentence 'Show the process of folding the square paper into a triangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of folding the square paper into a triangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 26.0], "base_se_norm": [-0.7729852199554443, -0.40976160764694214], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7956867218017578, -0.7048808336257935], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001792.pt"}
{"idx": 1794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcNcX0hEr_s.mp4", "duration": 88.1, "prompt_text": "Please find the visual event described by the sentence 'At what point is one of the top flaps folded down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is one of the top flaps folded down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 41.0], "base_se_norm": [-0.36435866355895996, -0.06923949718475342], "gt_se_raw": [55.0, 60.0], "gt_se_norm": [0.2485811710357666, 0.36208856105804443], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001794.pt"}
{"idx": 1796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcNcX0hEr_s.mp4", "duration": 88.1, "prompt_text": "Please find the visual event described by the sentence 'Show when the cup is being opened into its final 3D form.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show when the cup is being opened into its final 3D form.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 80.0], "base_se_norm": [0.6345062255859375, 0.8161181211471558], "gt_se_raw": [70.0, 78.0], "gt_se_norm": [0.5891033411026001, 0.7707151174545288], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001796.pt"}
{"idx": 1798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8jPbRnBwmXI.mp4", "duration": 67.93453333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a photo of a young girl is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a photo of a young girl is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 62.0], "base_se_norm": [0.3836851119995117, 0.8252867460250854], "gt_se_raw": [46.0, 62.0], "gt_se_norm": [0.3542449474334717, 0.8252867460250854], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001798.pt"}
{"idx": 1800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8jPbRnBwmXI.mp4", "duration": 67.93453333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention that the girl would take food from another person's plate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention that the girl would take food from another person's plate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.11679673194885254, -0.028476417064666748], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [0.23648452758789062, 0.29536473751068115], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001800.pt"}
{"idx": 1802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O7_TzpkCZrg.mp4", "duration": 45.07836666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a woman in a black shirt is seen talking in the playground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a woman in a black shirt is seen talking in the playground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 44.0], "base_se_norm": [-1.0, 0.9521559476852417], "gt_se_raw": [0.0, 45.0], "gt_se_norm": [-1.0, 0.9965231418609619], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001802.pt"}
{"idx": 1804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O7_TzpkCZrg.mp4", "duration": 45.07836666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman touch the beak of the yellow bird ride?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman touch the beak of the yellow bird ride?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.689429759979248, -0.5563281774520874], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.02392202615737915, 0.0648123025894165], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001804.pt"}
{"idx": 1806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O7_TzpkCZrg.mp4", "duration": 45.07836666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman uses both hands to gesture while describing the concept of the ride.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman uses both hands to gesture while describing the concept of the ride.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.46759384870529175, -0.2901250720024109], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [0.68595290184021, 0.8190544843673706], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001806.pt"}
{"idx": 1808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VD0zOrDEvAc.mp4", "duration": 189.05553333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the text '#RaceAnd RELIGION'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the text '#RaceAnd RELIGION'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.9365265965461731, -0.8942109942436218], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9259476661682129, -0.9047898650169373], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001808.pt"}
{"idx": 1810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VD0zOrDEvAc.mp4", "duration": 189.05553333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention the derogatory comments he receives, such as 'terrorist' and 'Osama'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention the derogatory comments he receives, such as 'terrorist' and 'Osama'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 97.0], "base_se_norm": [-0.09021437168121338, 0.026153564453125], "gt_se_raw": [87.0, 90.0], "gt_se_norm": [-0.07963550090789795, -0.04789876937866211], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001810.pt"}
{"idx": 1812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VZaqHyHFCzc.mp4", "duration": 150.76, "prompt_text": "Please find the visual event described by the sentence 'When can we see Sandman turning into a large amount of sand and rushing down the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Sandman turning into a large amount of sand and rushing down the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.8806049227714539], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.9469355344772339, -0.8806049227714539], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001812.pt"}
{"idx": 1814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VZaqHyHFCzc.mp4", "duration": 150.76, "prompt_text": "Please find the visual event described by the sentence 'When does Spider-Man fight Sandman inside the armored truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Spider-Man fight Sandman inside the armored truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 84.0], "base_se_norm": [-0.3499601483345032, 0.11435401439666748], "gt_se_raw": [49.0, 57.0], "gt_se_norm": [-0.3499601483345032, -0.24383121728897095], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001814.pt"}
{"idx": 1816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VZaqHyHFCzc.mp4", "duration": 150.76, "prompt_text": "Please find the visual event described by the sentence 'When is the armored truck flipped over during the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the armored truck flipped over during the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 60.0], "base_se_norm": [-0.25709736347198486, -0.20403289794921875], "gt_se_raw": [94.0, 98.0], "gt_se_norm": [0.247015118598938, 0.30007970333099365], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001816.pt"}
{"idx": 1818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i0Oj9HbsSCo.mp4", "duration": 9.609, "prompt_text": "Please find the visual event described by the sentence 'When does the girl drop the hairbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl drop the hairbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [0.4569673538208008, 0.8732438087463379], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.6651055812835693, 0.8732438087463379], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001818.pt"}
{"idx": 1820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w_y8z7jiBmk.mp4", "duration": 192.65913333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man start laughing hysterically at the cooking video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man start laughing hysterically at the cooking video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.43942445516586304, -0.3459951877593994], "gt_se_raw": [50.0, 60.0], "gt_se_norm": [-0.4809485673904419, -0.37713825702667236], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001820.pt"}
{"idx": 1822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w_y8z7jiBmk.mp4", "duration": 192.65913333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the show's title card 'OMKalen'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the show's title card 'OMKalen'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.730093240737915, -0.6885691285133362], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.7197122573852539, -0.6781880855560303], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001822.pt"}
{"idx": 1824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w_y8z7jiBmk.mp4", "duration": 192.65913333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the cooked chicken is being checked with a thermometer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 181.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the cooked chicken is being checked with a thermometer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 181.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 181.0], "base_se_norm": [0.7959179878234863, 0.878966212272644], "gt_se_raw": [177.0, 181.0], "gt_se_norm": [0.8374420404434204, 0.878966212272644], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001824.pt"}
{"idx": 1826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pGQd630EpLU.mp4", "duration": 13.833333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the person uses their right hand to wipe the stamps off the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the person uses their right hand to wipe the stamps off the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.5662650465965271, 0.15662658214569092], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.15662658214569092, 0.3012048006057739], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001826.pt"}
{"idx": 1828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pGQd630EpLU.mp4", "duration": 13.833333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the person point to the yellow sticker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person point to the yellow sticker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.4457831382751465, 0.7349398136138916], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [0.590361475944519, 0.7349398136138916], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001828.pt"}
{"idx": 1830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FOjYigV2FhA.mp4", "duration": 165.0649, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker introduce the topic of 'International Oddballs'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker introduce the topic of 'International Oddballs'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9394177794456482, -0.8788355588912964], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.927301287651062, -0.8909519910812378], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001830.pt"}
{"idx": 1832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FOjYigV2FhA.mp4", "duration": 165.0649, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man widens his eyes and raises his hands while talking about nuclear weapons power.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man widens his eyes and raises his hands while talking about nuclear weapons power.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [-0.4789928197860718, -0.4426434636116028], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.4668763279914856, -0.4305270314216614], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001832.pt"}
{"idx": 1834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FOjYigV2FhA.mp4", "duration": 165.0649, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the speaker lists the IMF and the World Bank as examples.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the speaker lists the IMF and the World Bank as examples.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 72.0], "base_se_norm": [-0.273013174533844, -0.12761586904525757], "gt_se_raw": [111.0, 113.0], "gt_se_norm": [0.34492552280426025, 0.3691585063934326], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001834.pt"}
{"idx": 1836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KVu2pLGGSNg.mp4", "duration": 175.375, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of the spider on its web.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of the spider on its web.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 31.0], "base_se_norm": [-0.7263007760047913, -0.6464718580245972], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7263007760047913, -0.6920883655548096], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001836.pt"}
{"idx": 1838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KVu2pLGGSNg.mp4", "duration": 175.375, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man kills the burnt zombie with a shovel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man kills the burnt zombie with a shovel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 106.0], "base_se_norm": [-0.12188166379928589, 0.20883822441101074], "gt_se_raw": [103.0, 106.0], "gt_se_norm": [0.17462575435638428, 0.20883822441101074], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001838.pt"}
{"idx": 1840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zVXT9iBoejQ.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When is the bar chart showing employment statistics for both disabled and non-disabled Americans displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the bar chart showing employment statistics for both disabled and non-disabled Americans displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001840.pt"}
{"idx": 1842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zVXT9iBoejQ.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the section of the video that presents the data for the year 1991.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the section of the video that presents the data for the year 1991.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001842.pt"}
{"idx": 1844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zVXT9iBoejQ.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame in which the 41.1% employment rate for disabled people in 2010 is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame in which the 41.1% employment rate for disabled people in 2010 is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [88, 91], "end_token_span": [87, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001844.pt"}
{"idx": 1846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iB52iP2a_MY.mp4", "duration": 95.12836666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the stuffed bunny picked up from the bowl and held up to the parrot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the stuffed bunny picked up from the bowl and held up to the parrot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 78.0], "base_se_norm": [0.4927198886871338, 0.6398894786834717], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.432345986366272, -0.36927330493927], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001846.pt"}
{"idx": 1848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iB52iP2a_MY.mp4", "duration": 95.12836666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the parrot is walking towards the right, away from the bunny.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the parrot is walking towards the right, away from the bunny.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 18.0], "base_se_norm": [-0.936927318572998, -0.6215639710426331], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.13530802726745605, 0.17735648155212402], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001848.pt"}
{"idx": 1850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X9nPJamttwM.mp4", "duration": 171.8717, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a staff member assisting a person in a wheelchair to board a train with a ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a staff member assisting a person in a wheelchair to board a train with a ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 130.0], "base_se_norm": [0.454573392868042, 0.5127562284469604], "gt_se_raw": [124.0, 129.0], "gt_se_norm": [0.4429367780685425, 0.5011197328567505], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001850.pt"}
{"idx": 1852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X9nPJamttwM.mp4", "duration": 171.8717, "prompt_text": "Please find the visual event described by the sentence 'At what point are people seen topping up their travel cards at the machines?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people seen topping up their travel cards at the machines?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 140.0], "base_se_norm": [0.582575798034668, 0.6291221380233765], "gt_se_raw": [137.0, 140.0], "gt_se_norm": [0.5942124128341675, 0.6291221380233765], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001852.pt"}
{"idx": 1854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X9nPJamttwM.mp4", "duration": 171.8717, "prompt_text": "Please find the visual event described by the sentence 'When is a visually impaired person with a guide dog shown getting on a bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a visually impaired person with a guide dog shown getting on a bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 169.0], "base_se_norm": [0.9316734075546265, 0.9665831327438354], "gt_se_raw": [166.0, 169.0], "gt_se_norm": [0.9316734075546265, 0.9665831327438354], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001854.pt"}
{"idx": 1856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xBD_s0RhUko.mp4", "duration": 180.63879166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man vacuum in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man vacuum in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 22.0], "base_se_norm": [-0.8117790818214417, -0.7564200162887573], "gt_se_raw": [18.0, 27.0], "gt_se_norm": [-0.8007072806358337, -0.7010608911514282], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001856.pt"}
{"idx": 1858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xBD_s0RhUko.mp4", "duration": 180.63879166666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man pick his nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man pick his nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [0.11825370788574219, 0.16254103183746338], "gt_se_raw": [102.0, 105.0], "gt_se_norm": [0.1293255090713501, 0.16254103183746338], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001858.pt"}
{"idx": 1860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uZ8lcB2IL6Q.mp4", "duration": 133.68, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where people are sitting on a bench overlooking the sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where people are sitting on a bench overlooking the sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 23.0], "base_se_norm": [-0.7606223821640015, -0.6558946371078491], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [-0.7606223821640015, -0.6708557605743408], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001860.pt"}
{"idx": 1862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uZ8lcB2IL6Q.mp4", "duration": 133.68, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a boy is playing a musical instrument.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a boy is playing a musical instrument.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 49.0], "base_se_norm": [-0.311789333820343, -0.2669060230255127], "gt_se_raw": [50.0, 55.0], "gt_se_norm": [-0.251944899559021, -0.17713940143585205], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001862.pt"}
{"idx": 1864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uZ8lcB2IL6Q.mp4", "duration": 133.68, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when men are rolling out a beach access mat on the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when men are rolling out a beach access mat on the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 105.0], "base_se_norm": [0.5260324478149414, 0.570915699005127], "gt_se_raw": [102.0, 105.0], "gt_se_norm": [0.5260324478149414, 0.570915699005127], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001864.pt"}
{"idx": 1866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yHsRjoRec8.mp4", "duration": 85.3853, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy puts the number 16 tile into its slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy puts the number 16 tile into its slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 50.0], "base_se_norm": [0.0774688720703125, 0.17116177082061768], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.3675726652145386, -0.3207261562347412], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001866.pt"}
{"idx": 1868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yHsRjoRec8.mp4", "duration": 85.3853, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman ask 'What comes before 11?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman ask 'What comes before 11?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.3207261562347412, -0.20361000299453735], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.45224058628082275, 0.4990870952606201], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001868.pt"}
{"idx": 1870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lozRy6VCLCI.mp4", "duration": 120.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the checkered shirt writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the checkered shirt writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 64.0], "base_se_norm": [-0.03429907560348511, 0.06560099124908447], "gt_se_raw": [58.0, 64.0], "gt_se_norm": [-0.03429907560348511, 0.06560099124908447], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001870.pt"}
{"idx": 1872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lozRy6VCLCI.mp4", "duration": 120.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when the choir members are walking in a circle around the chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when the choir members are walking in a circle around the chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 67.0], "base_se_norm": [0.06560099124908447, 0.11555111408233643], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.08225107192993164, 0.13220107555389404], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001872.pt"}
{"idx": 1874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lozRy6VCLCI.mp4", "duration": 120.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman with purple hair reading her sheet music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman with purple hair reading her sheet music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 116.0], "base_se_norm": [0.8814518451690674, 0.9314018487930298], "gt_se_raw": [112.0, 116.0], "gt_se_norm": [0.8648017644882202, 0.9314018487930298], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001874.pt"}
{"idx": 1876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zYX7iexkODw.mp4", "duration": 119.88643333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the father is putting a sock on the child's left foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the father is putting a sock on the child's left foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 44.0], "base_se_norm": [-0.6329860091209412, -0.2659720182418823], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.599621057510376, -0.46616148948669434], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001876.pt"}
{"idx": 1878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zYX7iexkODw.mp4", "duration": 119.88643333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the older brother walk over to bring the shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the older brother walk over to bring the shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 88.0], "base_se_norm": [0.3345963954925537, 0.46805596351623535], "gt_se_raw": [81.0, 87.0], "gt_se_norm": [0.35127878189086914, 0.4513735771179199], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001878.pt"}
{"idx": 1880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MqIJKnUkGLY.mp4", "duration": 237.94604166666664, "prompt_text": "Please find the visual event described by the sentence 'When does the man explain what makes a good parent?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man explain what makes a good parent?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 33.0], "base_se_norm": [-0.9327578544616699, -0.7226262092590332], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.9327578544616699, -0.8823262453079224], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001880.pt"}
{"idx": 1882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MqIJKnUkGLY.mp4", "duration": 237.94604166666664, "prompt_text": "Please find the visual event described by the sentence 'When is the lawyer with the beard and glasses shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the lawyer with the beard and glasses shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.6890051364898682, -0.6637893319129944], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [-0.1762838363647461, -0.14266276359558105], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001882.pt"}
{"idx": 1884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MqIJKnUkGLY.mp4", "duration": 237.94604166666664, "prompt_text": "Please find the visual event described by the sentence 'When does the man lose his temper and start banging on the witness stand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 219.0 - 221.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man lose his temper and start banging on the witness stand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 219.0 - 221.0 seconds.<|im_end|>\n", "base_se_raw": [219.0, 221.0], "base_se_norm": [0.840753436088562, 0.8575639724731445], "gt_se_raw": [219.0, 223.0], "gt_se_norm": [0.840753436088562, 0.874374508857727], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001884.pt"}
{"idx": 1886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z5UScMQUO6Q.mp4", "duration": 105.77233333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where someone is preparing canned tuna.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where someone is preparing canned tuna.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 50.0], "base_se_norm": [-0.14911586046218872, -0.0545731782913208], "gt_se_raw": [45.0, 50.0], "gt_se_norm": [-0.14911586046218872, -0.0545731782913208], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001886.pt"}
{"idx": 1888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z5UScMQUO6Q.mp4", "duration": 105.77233333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman is exercising at home with two children.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman is exercising at home with two children.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 84.0], "base_se_norm": [0.455957293510437, 0.5883170366287231], "gt_se_raw": [77.0, 84.0], "gt_se_norm": [0.455957293510437, 0.5883170366287231], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001888.pt"}
{"idx": 1890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdD8dLkEOAU.mp4", "duration": 126.70719520319788, "prompt_text": "Please find the visual event described by the sentence 'When does the girl open the 'Fun Unicorn' box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl open the 'Fun Unicorn' box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.9052934646606445, -0.8579401969909668], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8895090222358704, -0.8579401969909668], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001890.pt"}
{"idx": 1892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdD8dLkEOAU.mp4", "duration": 126.70719520319788, "prompt_text": "Please find the visual event described by the sentence 'At what point does the small toy transform into a large, magnificent winged unicorn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the small toy transform into a large, magnificent winged unicorn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [-0.4948984384536743, -0.415976345539093], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.4948984384536743, -0.4317607283592224], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001892.pt"}
{"idx": 1894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdD8dLkEOAU.mp4", "duration": 126.70719520319788, "prompt_text": "Please find the visual event described by the sentence 'When does the girl tickle the fat unicorn's belly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl tickle the fat unicorn's belly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [0.1838318109512329, 0.23118507862091064], "gt_se_raw": [74.0, 77.0], "gt_se_norm": [0.16804730892181396, 0.2154005765914917], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001894.pt"}
{"idx": 1896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfvyfN8ycvw.mp4", "duration": 61.93687499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a person running their hands through wood shavings in a black tray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a person running their hands through wood shavings in a black tray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.4510539770126343, -0.3218902349472046], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [-0.4833449125289917, -0.28959929943084717], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001896.pt"}
{"idx": 1898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfvyfN8ycvw.mp4", "duration": 61.93687499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a man in a yellow shirt is holding a long, thin yellow stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a man in a yellow shirt is holding a long, thin yellow stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 48.0], "base_se_norm": [0.1624739170074463, 0.5499651432037354], "gt_se_raw": [36.0, 48.0], "gt_se_norm": [0.1624739170074463, 0.5499651432037354], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001898.pt"}
{"idx": 1900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3BCIO1uX6f0.mp4", "duration": 181.4729583333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman eating food from a can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman eating food from a can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.8567279577255249], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.9448953866958618], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001900.pt"}
{"idx": 1902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3BCIO1uX6f0.mp4", "duration": 181.4729583333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man playing a game on the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man playing a game on the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.559162974357605, -0.4269118905067444], "gt_se_raw": [41.0, 51.0], "gt_se_norm": [-0.5481420755386353, -0.4379327893257141], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001902.pt"}
{"idx": 1904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3BCIO1uX6f0.mp4", "duration": 181.4729583333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man start repeating the word 'Minecraft'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start repeating the word 'Minecraft'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 114.0], "base_se_norm": [0.17923903465270996, 0.25638556480407715], "gt_se_raw": [127.0, 134.0], "gt_se_norm": [0.3996574878692627, 0.4768040180206299], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001904.pt"}
{"idx": 1906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HJENMThDg0k.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman waves her left hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman waves her left hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.635696530342102, -0.4656882882118225], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.635696530342102, -0.5628358125686646], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001906.pt"}
{"idx": 1908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HJENMThDg0k.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy have his hand near his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy have his hand near his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9271392822265625, -0.8299916982650757], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.10138481855392456, 0.020049691200256348], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001908.pt"}
{"idx": 1910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4gZpvVa0lis.mp4", "duration": 123.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the children doing sit-ups on the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children doing sit-ups on the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 16.0], "base_se_norm": [-0.9353622198104858, -0.7414489984512329], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.8545650243759155, -0.7576084136962891], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001910.pt"}
{"idx": 1912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4gZpvVa0lis.mp4", "duration": 123.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the dog lies on its back after switching places with the chimpanzee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the dog lies on its back after switching places with the chimpanzee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 83.0], "base_se_norm": [0.13116073608398438, 0.341233491897583], "gt_se_raw": [69.0, 79.0], "gt_se_norm": [0.11500132083892822, 0.27659571170806885], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001912.pt"}
{"idx": 1914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4gZpvVa0lis.mp4", "duration": 123.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the chimpanzee attempts to help the dog do a sit-up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the chimpanzee attempts to help the dog do a sit-up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 122.0], "base_se_norm": [0.13116073608398438, 0.9714516401290894], "gt_se_raw": [109.0, 113.0], "gt_se_norm": [0.7613788843154907, 0.8260166645050049], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001914.pt"}
{"idx": 1916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eRT_mIpXjbs.mp4", "duration": 385.52, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where 'Learning Science' is defined as the science of how people learn and its multidisciplinary sources are mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where 'Learning Science' is defined as the science of how people learn and its multidisciplinary sources are mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 65.0], "base_se_norm": [-0.8288026452064514, -0.6627930998802185], "gt_se_raw": [33.0, 52.0], "gt_se_norm": [-0.8288026452064514, -0.7302345037460327], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001916.pt"}
{"idx": 1918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eRT_mIpXjbs.mp4", "duration": 385.52, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the speaker explains how Learning Science helps identify non-verbal, automated knowledge from experts to make it available for training.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 184.0 - 202.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the speaker explains how Learning Science helps identify non-verbal, automated knowledge from experts to make it available for training.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 184.0 - 202.0 seconds.<|im_end|>\n", "base_se_raw": [184.0, 202.0], "base_se_norm": [-0.04544508457183838, 0.04793524742126465], "gt_se_raw": [183.0, 199.0], "gt_se_norm": [-0.05063289403915405, 0.0323718786239624], "start_token_span": [82, 87], "end_token_span": [89, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001918.pt"}
{"idx": 1920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LFoIjKYZtCY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where several individuals are engaged in welding activities together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where several individuals are engaged in welding activities together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.08288431167602539, 0.3756735324859619], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001920.pt"}
{"idx": 1922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LFoIjKYZtCY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where three people are gathered around a welding project.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where three people are gathered around a welding project.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001922.pt"}
{"idx": 1924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LFoIjKYZtCY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is the information about the program's duration displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the information about the program's duration displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.04813218116760254, 0.3756735324859619], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001924.pt"}
{"idx": 1926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p-_UlScFrQ8.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When are three men practicing first aid on a training dummy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three men practicing first aid on a training dummy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001926.pt"}
{"idx": 1928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p-_UlScFrQ8.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what time do two firefighters pose for the camera holding their helmets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do two firefighters pose for the camera holding their helmets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001928.pt"}
{"idx": 1930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G9YuKs3Jitk.mp4", "duration": 138.6385, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a plaid shirt being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a plaid shirt being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.8701659440994263, -0.7980359196662903], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.8701659440994263, -0.7980359196662903], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001930.pt"}
{"idx": 1932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G9YuKs3Jitk.mp4", "duration": 138.6385, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a woman and a child are looking at an animal through a wire fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a woman and a child are looking at an animal through a wire fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.3219776749610901, -0.2786996364593506], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.3219776749610901, -0.2786996364593506], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001932.pt"}
{"idx": 1934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G9YuKs3Jitk.mp4", "duration": 138.6385, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a boy is climbing a rock wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a boy is climbing a rock wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 86.0], "base_se_norm": [0.1973586082458496, 0.24063658714294434], "gt_se_raw": [83.0, 85.0], "gt_se_norm": [0.1973586082458496, 0.2262105941772461], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001934.pt"}
{"idx": 1936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICkM9rvBUzA.mp4", "duration": 135.135, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when chefs are preparing food in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when chefs are preparing food in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.7335997223854065, -0.6595996618270874], "gt_se_raw": [18.0, 23.0], "gt_se_norm": [-0.7335997223854065, -0.6595996618270874], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001936.pt"}
{"idx": 1938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICkM9rvBUzA.mp4", "duration": 135.135, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman named Sieglinde Alexander is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman named Sieglinde Alexander is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 63.0], "base_se_norm": [-0.23039919137954712, -0.06759905815124512], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.15639913082122803, -0.06759905815124512], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001938.pt"}
{"idx": 1940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFUE09SRO8w.mp4", "duration": 205.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'When are people participating in a pie-eating contest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people participating in a pie-eating contest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 23.0], "base_se_norm": [-0.8833630084991455, -0.7764458060264587], "gt_se_raw": [12.0, 24.0], "gt_se_norm": [-0.8833630084991455, -0.7667260766029358], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001940.pt"}
{"idx": 1942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFUE09SRO8w.mp4", "duration": 205.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the animation show a building on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the animation show a building on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [-0.19326096773147583, -0.17382144927978516], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [-0.20298069715499878, -0.18354123830795288], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001942.pt"}
{"idx": 1944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFUE09SRO8w.mp4", "duration": 205.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the microscopic view of chloroplasts moving within plant cells.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the microscopic view of chloroplasts moving within plant cells.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 138.0], "base_se_norm": [0.2635672092437744, 0.34132516384124756], "gt_se_raw": [130.0, 139.0], "gt_se_norm": [0.2635672092437744, 0.3510448932647705], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001944.pt"}
{"idx": 1946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMXsAtYtLTo.mp4", "duration": 104.5044, "prompt_text": "Please find the visual event described by the sentence 'When are tomatoes being washed on a conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are tomatoes being washed on a conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.6363789439201355, -0.5598271489143372], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.6363789439201355, -0.5789650678634644], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001946.pt"}
{"idx": 1948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMXsAtYtLTo.mp4", "duration": 104.5044, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the finished cans are being labeled and organized.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the finished cans are being labeled and organized.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.02396458387374878, 0.07172513008117676], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [-0.10051637887954712, -0.004826605319976807], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001948.pt"}
{"idx": 1950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iytz9nbqwi8.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when students are studying in a computer lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when students are studying in a computer lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.540437638759613], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001950.pt"}
{"idx": 1952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iytz9nbqwi8.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When is someone writing in a spiral notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is someone writing in a spiral notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08087527751922607, 0.3786870241165161], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001952.pt"}
{"idx": 1954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iytz9nbqwi8.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the text 'CAREERS' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the text 'CAREERS' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47478586435317993, -0.08087527751922607], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001954.pt"}
{"idx": 1956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Juj026QZGDo.mp4", "duration": 284.03375, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment where a woman with purple hair is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment where a woman with purple hair is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 94.0], "base_se_norm": [-0.46485233306884766, -0.33810681104660034], "gt_se_raw": [77.0, 94.0], "gt_se_norm": [-0.4578109383583069, -0.33810681104660034], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001956.pt"}
{"idx": 1958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Juj026QZGDo.mp4", "duration": 284.03375, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a man with a beard and a cap is shown petting a brown dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a man with a beard and a cap is shown petting a brown dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 189.0], "base_se_norm": [0.2744964361190796, 0.3308277130126953], "gt_se_raw": [181.0, 186.0], "gt_se_norm": [0.2744964361190796, 0.3097034692764282], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001958.pt"}
{"idx": 1960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s1FWVQFeOpQ.mp4", "duration": 130.06326666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman with purple hair speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman with purple hair speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 27.0], "base_se_norm": [-0.8769829273223877, -0.5848174095153809], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.8769829273223877, -0.8154743909835815], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001960.pt"}
{"idx": 1962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s1FWVQFeOpQ.mp4", "duration": 130.06326666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the black tank top talk about a pen that records lectures?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the black tank top talk about a pen that records lectures?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 46.0], "base_se_norm": [-0.4771775007247925, -0.292651891708374], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.38491469621658325, -0.3234061598777771], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001962.pt"}
{"idx": 1964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s1FWVQFeOpQ.mp4", "duration": 130.06326666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman is petting a black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman is petting a black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 97.0], "base_se_norm": [0.1379077434539795, 0.4915817975997925], "gt_se_raw": [74.0, 81.0], "gt_se_norm": [0.1379077434539795, 0.24554777145385742], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001964.pt"}
{"idx": 1966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymc30meWzfg.mp4", "duration": 78.74533333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is a man's body composition being measured while he is submerged in a pool of water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a man's body composition being measured while he is submerged in a pool of water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 42.0], "base_se_norm": [-0.3142450451850891, 0.06672990322113037], "gt_se_raw": [26.0, 42.0], "gt_se_norm": [-0.33964335918426514, 0.06672990322113037], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001966.pt"}
{"idx": 1968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymc30meWzfg.mp4", "duration": 78.74533333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a doctor is using an ultrasound machine to examine a patient's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a doctor is using an ultrasound machine to examine a patient's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 55.0], "base_se_norm": [0.14292490482330322, 0.396908164024353], "gt_se_raw": [45.0, 56.0], "gt_se_norm": [0.14292490482330322, 0.42230653762817383], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001968.pt"}
{"idx": 1970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YOqmroV2cRo.mp4", "duration": 66.7667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the child looking at the toy board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the child looking at the toy board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 15.0], "base_se_norm": [-0.880179762840271, -0.5506742000579834], "gt_se_raw": [4.0, 14.0], "gt_se_norm": [-0.880179762840271, -0.5806292295455933], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001970.pt"}
{"idx": 1972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YOqmroV2cRo.mp4", "duration": 66.7667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child touch her ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child touch her ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.43085402250289917, -0.3409888744354248], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.43085402250289917, -0.37094390392303467], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001972.pt"}
{"idx": 1974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YOqmroV2cRo.mp4", "duration": 66.7667, "prompt_text": "Please find the visual event described by the sentence 'When is a woman visible walking in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman visible walking in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9101348519325256], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.13130348920822144, -0.04143834114074707], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001974.pt"}
{"idx": 1976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSjK0X2XLxY.mp4", "duration": 191.52485819152486, "prompt_text": "Please find the visual event described by the sentence 'Show the process of mixing water, washing up liquid, and glycerine in a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of mixing water, washing up liquid, and glycerine in a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 98.0], "base_se_norm": [-0.32123690843582153, 0.02336585521697998], "gt_se_raw": [64.0, 94.0], "gt_se_norm": [-0.33167946338653564, -0.01840418577194214], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001976.pt"}
{"idx": 1978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSjK0X2XLxY.mp4", "duration": 191.52485819152486, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot where the bubble exhibits swirling rainbow-colored patterns.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot where the bubble exhibits swirling rainbow-colored patterns.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 155.0], "base_se_norm": [0.3784111738204956, 0.6185888051986694], "gt_se_raw": [133.0, 155.0], "gt_se_norm": [0.38885366916656494, 0.6185888051986694], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001978.pt"}
{"idx": 1980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM-e5QkRNo.mp4", "duration": 27.39267296229322, "prompt_text": "Please find the visual event described by the sentence 'When is the dog actively jumping up and down against the gate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog actively jumping up and down against the gate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, 0.38723230361938477], "gt_se_raw": [0.0, 18.0], "gt_se_norm": [-1.0, 0.3142200708389282], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001980.pt"}
{"idx": 1982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM-e5QkRNo.mp4", "duration": 27.39267296229322, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog give up and lie down on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog give up and lie down on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.38723230361938477, 0.6062688827514648], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.38723230361938477, 0.6062688827514648], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001982.pt"}
{"idx": 1984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM-e5QkRNo.mp4", "duration": 27.39267296229322, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the dog is wearing a bandana?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the dog is wearing a bandana?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 27.0], "base_se_norm": [-1.0, 0.9713300466537476], "gt_se_raw": [0.0, 27.0], "gt_se_norm": [-1.0, 0.9713300466537476], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001984.pt"}
{"idx": 1986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r07eYlcQLJQ.mp4", "duration": 70.1, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the camera focuses on the fire pit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the camera focuses on the fire pit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 29.0], "base_se_norm": [-0.343794584274292, -0.17261052131652832], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.2867332100868225, -0.20114123821258545], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001986.pt"}
{"idx": 1988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r07eYlcQLJQ.mp4", "duration": 70.1, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that provides a clear view of the field enclosed by a white fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that provides a clear view of the field enclosed by a white fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 64.0], "base_se_norm": [-0.0014265179634094238, 0.8259629011154175], "gt_se_raw": [39.0, 47.0], "gt_se_norm": [0.11269617080688477, 0.34094154834747314], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001988.pt"}
{"idx": 1990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1ycCh8MGLDo.mp4", "duration": 269.72779166666663, "prompt_text": "Please find the visual event described by the sentence 'When are the tapioca pearls poured into the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the tapioca pearls poured into the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.8665320873260498, -0.8072130680084229], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.8591172099113464, -0.8220428228378296], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001990.pt"}
{"idx": 1992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1ycCh8MGLDo.mp4", "duration": 269.72779166666663, "prompt_text": "Please find the visual event described by the sentence 'At what point is the soy milk poured into the pot with the tea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the soy milk poured into the pot with the tea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 111.0], "base_se_norm": [-0.21402239799499512, -0.1769479513168335], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [-0.20660752058029175, -0.19177770614624023], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001992.pt"}
{"idx": 1994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1ycCh8MGLDo.mp4", "duration": 269.72779166666663, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the cooked tapioca pearls are put into the glass bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 211.0 - 229.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the cooked tapioca pearls are put into the glass bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 211.0 - 229.0 seconds.<|im_end|>\n", "base_se_raw": [211.0, 229.0], "base_se_norm": [0.5645403861999512, 0.6980082988739014], "gt_se_raw": [215.0, 229.0], "gt_se_norm": [0.5941998958587646, 0.6980082988739014], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001994.pt"}
{"idx": 1996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vY0I6yjBy1c.mp4", "duration": 92.42566666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the first long shot of the empty grocery store aisle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the first long shot of the empty grocery store aisle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.04788351058959961, 0.038672566413879395], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.04788351058959961, 0.01703345775604248], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001996.pt"}
{"idx": 1998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vY0I6yjBy1c.mp4", "duration": 92.42566666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up of the sign that reads 'Thank you for shopping with others in mind.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up of the sign that reads 'Thank you for shopping with others in mind.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.2767016887664795, 0.36325764656066895], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.2767016887664795, 0.36325764656066895], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001998.pt"}
{"idx": 2000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4ClKFnnzSRA.mp4", "duration": 134.86806666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the purple shirt first appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the purple shirt first appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9555121064186096, -0.8665362596511841], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9703413844108582, -0.9406827688217163], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002000.pt"}
{"idx": 2002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4ClKFnnzSRA.mp4", "duration": 134.86806666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a person in the background leave the room and close the door behind them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a person in the background leave the room and close the door behind them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 87.0], "base_se_norm": [0.24566173553466797, 0.2901496887207031], "gt_se_raw": [87.0, 90.0], "gt_se_norm": [0.2901496887207031, 0.3346376419067383], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002002.pt"}
{"idx": 2004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4ClKFnnzSRA.mp4", "duration": 134.86806666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in purple starts writing on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in purple starts writing on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 134.0], "base_se_norm": [0.868492603302002, 0.9871270656585693], "gt_se_raw": [120.0, 123.0], "gt_se_norm": [0.7795168161392212, 0.8240047693252563], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002004.pt"}
{"idx": 2006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mnd7sFt5c3A.mp4", "duration": 156.82333333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a boombox is shown up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a boombox is shown up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9744935631752014], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9744935631752014], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002006.pt"}
{"idx": 2008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mnd7sFt5c3A.mp4", "duration": 156.82333333333332, "prompt_text": "Please find the visual event described by the sentence 'Where can we see people in white hazmat suits exploring a strange place?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see people in white hazmat suits exploring a strange place?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 95.0], "base_se_norm": [0.1732947826385498, 0.2115544080734253], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [0.1732947826385498, 0.2115544080734253], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002008.pt"}
{"idx": 2010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V49ENdZlOx4.mp4", "duration": 30.0, "prompt_text": "Please find the visual event described by the sentence 'When do the two minions look bored at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two minions look bored at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.7333333492279053], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.800000011920929], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002010.pt"}
{"idx": 2012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V49ENdZlOx4.mp4", "duration": 30.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does one minion throw a pair of red boxing gloves onto the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one minion throw a pair of red boxing gloves onto the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 14.0], "base_se_norm": [-0.13333332538604736, -0.06666666269302368], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.19999998807907104, -0.06666666269302368], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002012.pt"}
{"idx": 2014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V49ENdZlOx4.mp4", "duration": 30.0, "prompt_text": "Please find the visual event described by the sentence 'When does the minion point towards the other minion with a surprised expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the minion point towards the other minion with a surprised expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [0.7999999523162842, 0.9333332777023315], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.7999999523162842, 0.9333332777023315], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002014.pt"}
{"idx": 2016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/duF40iZq464.mp4", "duration": 134.50103333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the researcher write notes in her notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the researcher write notes in her notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.44981837272644043, -0.3605997562408447], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.44981837272644043, -0.37546950578689575], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002016.pt"}
{"idx": 2018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/duF40iZq464.mp4", "duration": 134.50103333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when an owl is perched on a branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when an owl is perched on a branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 90.0], "base_se_norm": [0.29367029666900635, 0.3382796049118042], "gt_se_raw": [87.0, 90.0], "gt_se_norm": [0.29367029666900635, 0.3382796049118042], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002018.pt"}
{"idx": 2020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-cf_-i_gCdY.mp4", "duration": 138.5, "prompt_text": "Please find the visual event described by the sentence 'A woman reads a book titled 'Where's Waldo?' while sitting in an office chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman reads a book titled 'Where's Waldo?' while sitting in an office chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.9133573770523071], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.9277978539466858], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002020.pt"}
{"idx": 2022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-cf_-i_gCdY.mp4", "duration": 138.5, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man reading a book while submerged in a swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man reading a book while submerged in a swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [-0.46570396423339844, -0.4079422354698181], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.46570396423339844, -0.43682312965393066], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002022.pt"}
{"idx": 2024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-cf_-i_gCdY.mp4", "duration": 138.5, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a green t-shirt speaking directly to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a green t-shirt speaking directly to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 84.0], "base_se_norm": [0.12635374069213867, 0.21299636363983154], "gt_se_raw": [77.0, 85.0], "gt_se_norm": [0.11191332340240479, 0.22743678092956543], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002024.pt"}
{"idx": 2026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8eBmqNyxBg.mp4", "duration": 117.7176, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the chicken slaps the woman's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the chicken slaps the woman's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 50.0], "base_se_norm": [-0.1844889521598816, -0.15050935745239258], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.1674991250038147, -0.13351953029632568], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002026.pt"}
{"idx": 2028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8eBmqNyxBg.mp4", "duration": 117.7176, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman looks at the baby monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman looks at the baby monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 88.0], "base_se_norm": [0.4441341161727905, 0.4951035976409912], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.4441341161727905, 0.4781137704849243], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002028.pt"}
{"idx": 2030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GoDvDNGaPUU.mp4", "duration": 15.081733333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the parrot peers into the cardboard box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the parrot peers into the cardboard box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.6021677255630493, 0.7239397764205933], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.7347785234451294, -0.469556987285614], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002030.pt"}
{"idx": 2032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GoDvDNGaPUU.mp4", "duration": 15.081733333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we hear the parrot speak for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we hear the parrot speak for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.3261075019836426, 0.5913289785385132], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [0.1934967041015625, 0.3261075019836426], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002032.pt"}
{"idx": 2034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GoDvDNGaPUU.mp4", "duration": 15.081733333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the parrot's long tail feathers are hanging off the side of the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the parrot's long tail feathers are hanging off the side of the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 14.0], "base_se_norm": [-0.8673892617225647, 0.8565504550933838], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.8673892617225647, 0.9891612529754639], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002034.pt"}
{"idx": 2036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kO_mpRhI1jY.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does an instructor point to a projector screen in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does an instructor point to a projector screen in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.540437638759613, -0.08087527751922607], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002036.pt"}
{"idx": 2038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kO_mpRhI1jY.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a young man in a grey t-shirt is looking at a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a young man in a grey t-shirt is looking at a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4443387985229492, 0.8382494449615479], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3786870241165161, 0.8382494449615479], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002038.pt"}
{"idx": 2040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UkcYowvTRdA.mp4", "duration": 55.7, "prompt_text": "Please find the visual event described by the sentence 'When are the company's logos shown on a white background at the start of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the company's logos shown on a white background at the start of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9281867146492004], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9281867146492004], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002040.pt"}
{"idx": 2042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UkcYowvTRdA.mp4", "duration": 55.7, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman with straight brown hair is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman with straight brown hair is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 43.0], "base_se_norm": [-0.17414724826812744, 0.5439856052398682], "gt_se_raw": [23.0, 43.0], "gt_se_norm": [-0.17414724826812744, 0.5439856052398682], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002042.pt"}
{"idx": 2044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UkcYowvTRdA.mp4", "duration": 55.7, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when two women are talking in front of a backdrop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when two women are talking in front of a backdrop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 50.0], "base_se_norm": [-0.9281867146492004, 0.7953320741653442], "gt_se_raw": [2.0, 50.0], "gt_se_norm": [-0.9281867146492004, 0.7953320741653442], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002044.pt"}
{"idx": 2046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FeaDpI96lIA.mp4", "duration": 67.20045045045045, "prompt_text": "Please find the visual event described by the sentence 'Show the part where four people in aprons are smiling at the camera in a kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where four people in aprons are smiling at the camera in a kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.6130977869033813, -0.5238126516342163], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.6428595185279846, -0.5238126516342163], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002046.pt"}
{"idx": 2048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FeaDpI96lIA.mp4", "duration": 67.20045045045045, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a woman in a classroom raising her hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a woman in a classroom raising her hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 33.0], "base_se_norm": [-0.07738709449768066, -0.017863690853118896], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [-0.07738709449768066, -0.017863690853118896], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002048.pt"}
{"idx": 2050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VqrBsMFRaLA.mp4", "duration": 203.11957777777778, "prompt_text": "Please find the visual event described by the sentence 'When does the character dressed as Jafar appear, holding his staff and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character dressed as Jafar appear, holding his staff and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 77.0], "base_se_norm": [-0.4682934880256653, -0.24182593822479248], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [-0.4781399369239807, -0.448600709438324], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002050.pt"}
{"idx": 2052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VqrBsMFRaLA.mp4", "duration": 203.11957777777778, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man with blue hair and a skull on his shoulder playing with chess pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man with blue hair and a skull on his shoulder playing with chess pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 45.0], "base_se_norm": [-0.6258361339569092, -0.5569112300872803], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.6258361339569092, -0.5864505171775818], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002052.pt"}
{"idx": 2054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VqrBsMFRaLA.mp4", "duration": 203.11957777777778, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the woman with white hair is singing with a shell necklace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the woman with white hair is singing with a shell necklace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 35.0], "base_se_norm": [-0.6750682592391968, -0.6553754210472107], "gt_se_raw": [133.0, 136.0], "gt_se_norm": [0.3095734119415283, 0.33911263942718506], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002054.pt"}
{"idx": 2056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KHHrCdQZ5dk.mp4", "duration": 30.2, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl gets up from the blue mat and runs towards the green steps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl gets up from the blue mat and runs towards the green steps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 28.0], "base_se_norm": [-0.0066225528717041016, 0.8543045520782471], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [0.65562903881073, 0.8543045520782471], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002056.pt"}
{"idx": 2058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KHHrCdQZ5dk.mp4", "duration": 30.2, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the little boy in the orange t-shirt is touching the side of the red ramp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the little boy in the orange t-shirt is touching the side of the red ramp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.8675496578216553, -0.6026489734649658], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, -0.20529800653457642], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002058.pt"}
{"idx": 2060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vKtnj21k1Cg.mp4", "duration": 129.7296, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is typing on a Braille keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is typing on a Braille keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.7841664552688599, -0.7379164099693298], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.7995831370353699, -0.7379164099693298], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002060.pt"}
{"idx": 2062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vKtnj21k1Cg.mp4", "duration": 129.7296, "prompt_text": "Please find the visual event described by the sentence 'At what moment is a woman with short grey hair and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is a woman with short grey hair and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 68.0], "base_se_norm": [-0.16749918460845947, 0.048334360122680664], "gt_se_raw": [55.0, 68.0], "gt_se_norm": [-0.15208250284194946, 0.048334360122680664], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002062.pt"}
{"idx": 2064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vKtnj21k1Cg.mp4", "duration": 129.7296, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a man is demonstrating a cash detector with a dollar bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a man is demonstrating a cash detector with a dollar bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 101.0], "base_se_norm": [0.4491680860519409, 0.5570849180221558], "gt_se_raw": [95.0, 100.0], "gt_se_norm": [0.4645848274230957, 0.541668176651001], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002064.pt"}
{"idx": 2066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gRD53bcAM8E.mp4", "duration": 52.11873333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the shrimp-like creature is texting on a phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the shrimp-like creature is texting on a phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.3860173225402832, -0.1557738184928894], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.4243912100791931, -0.1557738184928894], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002066.pt"}
{"idx": 2068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gRD53bcAM8E.mp4", "duration": 52.11873333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the red monster starts playing the drums.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the red monster starts playing the drums.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [0.0744696855545044, 0.2279653549194336], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.0744696855545044, 0.15121757984161377], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002068.pt"}
{"idx": 2070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7jck10cm5g.mp4", "duration": 98.16473333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a person holding up a 'Batman' comic book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person holding up a 'Batman' comic book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9592521786689758], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9796260595321655, -0.9592521786689758], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002070.pt"}
{"idx": 2072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7jck10cm5g.mp4", "duration": 98.16473333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the wide shot showing a large, multi-level library filled with people at tables.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the wide shot showing a large, multi-level library filled with people at tables.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.34803467988967896, -0.18504339456558228], "gt_se_raw": [31.0, 40.0], "gt_se_norm": [-0.3684086203575134, -0.18504339456558228], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002072.pt"}
{"idx": 2074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7jck10cm5g.mp4", "duration": 98.16473333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a computer screen displays the Vision Australia library website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a computer screen displays the Vision Australia library website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 90.0], "base_se_norm": [0.7114088535308838, 0.8336523771286011], "gt_se_raw": [84.0, 90.0], "gt_se_norm": [0.7114088535308838, 0.8336523771286011], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002074.pt"}
{"idx": 2076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGeKSiCQkPw.mp4", "duration": 80.6806, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the dog turns its head to the side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the dog turns its head to the side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.33069413900375366, -0.2067486047744751], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.2563268542289734, -0.18195950984954834], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002076.pt"}
{"idx": 2078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGeKSiCQkPw.mp4", "duration": 80.6806, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the dog's eyes are wide and it's staring directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the dog's eyes are wide and it's staring directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.8512653708457947], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [0.21466612815856934, 0.3386117219924927], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002078.pt"}
{"idx": 2080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wkbyROy79oY.mp4", "duration": 107.2071, "prompt_text": "Please find the visual event described by the sentence 'When is the woman explaining that her son was born with hydrocephalus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman explaining that her son was born with hydrocephalus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 17.0], "base_se_norm": [-0.9253780841827393, -0.6828567981719971], "gt_se_raw": [18.0, 26.0], "gt_se_norm": [-0.6642013788223267, -0.5149574875831604], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002080.pt"}
{"idx": 2082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wkbyROy79oY.mp4", "duration": 107.2071, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is a pi\u00f1ata shaped like a truck visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is a pi\u00f1ata shaped like a truck visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [0.26857268810272217, 0.36185014247894287], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [0.26857268810272217, 0.36185014247894287], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002082.pt"}
{"idx": 2084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wkbyROy79oY.mp4", "duration": 107.2071, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a teenager is stirring a red liquid in a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a teenager is stirring a red liquid in a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 106.0], "base_se_norm": [0.8842035531997681, 0.9774810075759888], "gt_se_raw": [101.0, 107.0], "gt_se_norm": [0.8842035531997681, 0.9961364269256592], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002084.pt"}
{"idx": 2086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ocAL0oz3HxU.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy, wearing a striped shirt, plays with a purple coiled toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy, wearing a striped shirt, plays with a purple coiled toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 63.0], "base_se_norm": [-0.5827561616897583, -0.24896103143692017], "gt_se_raw": [36.0, 60.0], "gt_se_norm": [-0.5708348751068115, -0.28472477197647095], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002086.pt"}
{"idx": 2088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ocAL0oz3HxU.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy is standing and interacting with various hanging items, including a pinwheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy is standing and interacting with various hanging items, including a pinwheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 102.0], "base_se_norm": [-0.04629969596862793, 0.21596789360046387], "gt_se_raw": [80.0, 102.0], "gt_se_norm": [-0.04629969596862793, 0.21596789360046387], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002088.pt"}
{"idx": 2090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0x1EwpvvyeQ.mp4", "duration": 120.38693333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman in a wheelchair is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman in a wheelchair is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 73.0], "base_se_norm": [0.09646451473236084, 0.21275615692138672], "gt_se_raw": [67.0, 73.0], "gt_se_norm": [0.11307764053344727, 0.21275615692138672], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002090.pt"}
{"idx": 2092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0x1EwpvvyeQ.mp4", "duration": 120.38693333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video demonstrate the integrated voice control feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video demonstrate the integrated voice control feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [0.030012130737304688, 0.09646451473236084], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [0.030012130737304688, 0.09646451473236084], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002092.pt"}
{"idx": 2094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0x1EwpvvyeQ.mp4", "duration": 120.38693333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the 'The Sesame Phone' title card is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the 'The Sesame Phone' title card is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.7009642124176025, -0.6511249542236328], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.717577338218689, -0.6511249542236328], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002094.pt"}
{"idx": 2096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L9KC5W7A2yI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two men in plaid shirts are inspecting the internal components of a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two men in plaid shirts are inspecting the internal components of a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002096.pt"}
{"idx": 2098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L9KC5W7A2yI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time a woman is seen standing next to an air conditioning unit holding gauges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time a woman is seen standing next to an air conditioning unit holding gauges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002098.pt"}
{"idx": 2100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfyJQEIsMt0.mp4", "duration": 319.3607083333333, "prompt_text": "Please find the visual event described by the sentence 'When does the grey cloud create a baby alligator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the grey cloud create a baby alligator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [-0.605461835861206, -0.5866742730140686], "gt_se_raw": [59.0, 64.0], "gt_se_norm": [-0.6305118799209595, -0.5991992950439453], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002100.pt"}
{"idx": 2102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfyJQEIsMt0.mp4", "duration": 319.3607083333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a porcupine is created by the cloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a porcupine is created by the cloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 168.0], "base_se_norm": [0.0020017623901367188, 0.052101850509643555], "gt_se_raw": [159.0, 161.0], "gt_se_norm": [-0.0042607784271240234, 0.008264303207397461], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002102.pt"}
{"idx": 2104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfyJQEIsMt0.mp4", "duration": 319.3607083333333, "prompt_text": "Please find the visual event described by the sentence 'When does the stork put on protective football gear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 249.0 - 254.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the stork put on protective football gear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 249.0 - 254.0 seconds.<|im_end|>\n", "base_se_raw": [249.0, 254.0], "base_se_norm": [0.5593652725219727, 0.5906777381896973], "gt_se_raw": [249.0, 253.0], "gt_se_norm": [0.5593652725219727, 0.5844153165817261], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002104.pt"}
{"idx": 2106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FkTybqcX-Yo.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'When does the man jump off the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man jump off the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 35.0], "base_se_norm": [-0.5499380826950073, -0.5226616263389587], "gt_se_raw": [34.0, 35.0], "gt_se_norm": [-0.5362998843193054, -0.5226616263389587], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002106.pt"}
{"idx": 2108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FkTybqcX-Yo.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'When does Captain America hold onto the helicopter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Captain America hold onto the helicopter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [-0.39991748332977295, -0.3590027689933777], "gt_se_raw": [116.0, 117.0], "gt_se_norm": [0.5820356607437134, 0.5956739187240601], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002108.pt"}
{"idx": 2110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GaoLU6zKaws.mp4", "duration": 302.5, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man playing the saxophone in a laundromat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man playing the saxophone in a laundromat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.8148760199546814, -0.7884297370910645], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.8148760199546814, -0.7884297370910645], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002110.pt"}
{"idx": 2112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GaoLU6zKaws.mp4", "duration": 302.5, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man is playing the saxophone on a high school football field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 178.0 - 192.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man is playing the saxophone on a high school football field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 178.0 - 192.0 seconds.<|im_end|>\n", "base_se_raw": [178.0, 192.0], "base_se_norm": [0.17685949802398682, 0.26942145824432373], "gt_se_raw": [178.0, 208.0], "gt_se_norm": [0.17685949802398682, 0.3752065896987915], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002112.pt"}
{"idx": 2114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GaoLU6zKaws.mp4", "duration": 302.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the performer is playing the saxophone for students during a class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the performer is playing the saxophone for students during a class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 162.0], "base_se_norm": [-0.16033059358596802, 0.07107436656951904], "gt_se_raw": [127.0, 162.0], "gt_se_norm": [-0.16033059358596802, 0.07107436656951904], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002114.pt"}
{"idx": 2116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vjW4UOC7U3w.mp4", "duration": 73.30656666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see two boys in a toy race car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two boys in a toy race car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.12695401906967163, -0.04510599374771118], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.12695401906967163, -0.04510599374771118], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002116.pt"}
{"idx": 2118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vjW4UOC7U3w.mp4", "duration": 73.30656666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the Mamba roller coaster shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the Mamba roller coaster shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 65.0], "base_se_norm": [0.44598233699798584, 0.7733745574951172], "gt_se_raw": [52.0, 65.0], "gt_se_norm": [0.4186997413635254, 0.7733745574951172], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002118.pt"}
{"idx": 2120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2PjZAeiU7uM.mp4", "duration": 154.98816666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with red hair answer the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with red hair answer the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9612873792648315, -0.9225747585296631], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.9612873792648315, -0.9225747585296631], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002120.pt"}
{"idx": 2122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2PjZAeiU7uM.mp4", "duration": 154.98816666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman get out of the car wearing red high heels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman get out of the car wearing red high heels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [-0.5225441455841064, -0.47092729806900024], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.5225441455841064, -0.47092729806900024], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002122.pt"}
{"idx": 2124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2PjZAeiU7uM.mp4", "duration": 154.98816666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with white hair arrive in the office and take off her sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with white hair arrive in the office and take off her sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 99.0], "base_se_norm": [0.2259000539779663, 0.2775169610977173], "gt_se_raw": [95.0, 98.0], "gt_se_norm": [0.2259000539779663, 0.2646127939224243], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002124.pt"}
{"idx": 2126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YipS4o9EaoU.mp4", "duration": 158.65849182515848, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the various symptoms of ME/CFS are listed on a light blue background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the various symptoms of ME/CFS are listed on a light blue background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.7478861808776855, -0.647040605545044], "gt_se_raw": [20.0, 28.0], "gt_se_norm": [-0.7478861808776855, -0.647040605545044], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002126.pt"}
{"idx": 2128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YipS4o9EaoU.mp4", "duration": 158.65849182515848, "prompt_text": "Please find the visual event described by the sentence 'When does the video discuss the potential causes, or 'Aetiology', of the illness?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video discuss the potential causes, or 'Aetiology', of the illness?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 56.0], "base_se_norm": [-0.4957723617553711, -0.29408127069473267], "gt_se_raw": [41.0, 56.0], "gt_se_norm": [-0.4831666350364685, -0.29408127069473267], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002128.pt"}
{"idx": 2130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MCKeh7ynQjQ.mp4", "duration": 98.01458333333332, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the speaker state that graduation day is her favorite day of the year?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the speaker state that graduation day is her favorite day of the year?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [-0.32663077116012573, -0.2654154300689697], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8163538575172424, -0.7347333431243896], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002130.pt"}
{"idx": 2132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MCKeh7ynQjQ.mp4", "duration": 98.01458333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me when she announces the specific new date reserved for the ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when she announces the specific new date reserved for the ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [0.1630922555923462, 0.265117883682251], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [0.1630922555923462, 0.22430765628814697], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002132.pt"}
{"idx": 2134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MCKeh7ynQjQ.mp4", "duration": 98.01458333333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the speaker puts on her graduation cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the speaker puts on her graduation cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 91.0], "base_se_norm": [0.7548409700393677, 0.8568665981292725], "gt_se_raw": [87.0, 90.0], "gt_se_norm": [0.7752461433410645, 0.8364614248275757], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002134.pt"}
{"idx": 2136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1DDhUnyvwY.mp4", "duration": 162.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the camera look up at the sky through the tree canopy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the camera look up at the sky through the tree canopy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 49.0], "base_se_norm": [-0.520295262336731, -0.39729398488998413], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.520295262336731, -0.4710947275161743], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002136.pt"}
{"idx": 2138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1DDhUnyvwY.mp4", "duration": 162.6, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a green snake is slithering on a mossy branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a green snake is slithering on a mossy branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 104.0], "base_se_norm": [0.24231243133544922, 0.27921271324157715], "gt_se_raw": [102.0, 104.0], "gt_se_norm": [0.25461244583129883, 0.27921271324157715], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002138.pt"}
{"idx": 2140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iRsB9WrSePA.mp4", "duration": 75.44210877544211, "prompt_text": "Please find the visual event described by the sentence 'When is a man in a teal shirt seen talking on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man in a teal shirt seen talking on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.5228129029273987, -0.4167713522911072], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.5228129029273987, -0.4167713522911072], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002140.pt"}
{"idx": 2142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iRsB9WrSePA.mp4", "duration": 75.44210877544211, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman in a red shirt is speaking to a man standing next to her desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman in a red shirt is speaking to a man standing next to her desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [0.35203003883361816, 0.4845820665359497], "gt_se_raw": [50.0, 56.0], "gt_se_norm": [0.3255196809768677, 0.4845820665359497], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002142.pt"}
{"idx": 2144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iRsB9WrSePA.mp4", "duration": 75.44210877544211, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a woman is looking at some papers at her desk with two monitors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a woman is looking at some papers at her desk with two monitors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.25770896673202515, -0.15166741609573364], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.25770896673202515, -0.15166741609573364], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002144.pt"}
{"idx": 2146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SY0VyoUP5uw.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show the 41.1% employment rate for disabled Americans in 2010?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show the 41.1% employment rate for disabled Americans in 2010?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [86, 89], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002146.pt"}
{"idx": 2148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SY0VyoUP5uw.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show me the bar chart that illustrates the employment gap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the bar chart that illustrates the employment gap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [71, 74], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002148.pt"}
{"idx": 2150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OzP8k0R-USw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is coughing in bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is coughing in bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9505138993263245], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.9628854393959045, -0.9010277390480042], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002150.pt"}
{"idx": 2152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OzP8k0R-USw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the blue suit giving a speech to the crowd?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the blue suit giving a speech to the crowd?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 98.0], "base_se_norm": [-0.3566804528236389, 0.2124098539352417], "gt_se_raw": [51.0, 60.0], "gt_se_norm": [-0.369051992893219, -0.25770825147628784], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002152.pt"}
{"idx": 2154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OzP8k0R-USw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the crowd makes a toast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the crowd makes a toast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 131.0], "base_se_norm": [0.261896014213562, 0.6206703186035156], "gt_se_raw": [129.0, 131.0], "gt_se_norm": [0.595927357673645, 0.6206703186035156], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002154.pt"}
{"idx": 2156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MrFs4lL2pbc.mp4", "duration": 91.32456666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the children's hands interlaced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the children's hands interlaced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 61.0], "base_se_norm": [-0.5620017647743225, 0.3358945846557617], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.4963020086288452, -0.40870237350463867], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002156.pt"}
{"idx": 2158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MrFs4lL2pbc.mp4", "duration": 91.32456666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the girl in the purple shirt says 'This is fun'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the girl in the purple shirt says 'This is fun'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 73.0], "base_se_norm": [0.5548937320709229, 0.5986934900283813], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.8248007297515869, -0.7810008525848389], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002158.pt"}
{"idx": 2160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzaVd6zl2bA.mp4", "duration": 106.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the black hood playing the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the black hood playing the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.6997184753417969, -0.5871129035949707], "gt_se_raw": [1.0, 17.0], "gt_se_norm": [-0.9812324047088623, -0.6809508800506592], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002160.pt"}
{"idx": 2162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzaVd6zl2bA.mp4", "duration": 106.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man pull open the curtains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man pull open the curtains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 45.0], "base_se_norm": [-0.2492961883544922, -0.1554582118988037], "gt_se_raw": [40.0, 41.0], "gt_se_norm": [-0.2492961883544922, -0.2305285930633545], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002162.pt"}
{"idx": 2164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzaVd6zl2bA.mp4", "duration": 106.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'When do the two men start dancing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two men start dancing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 105.0], "base_se_norm": [0.782921552658081, 0.970597505569458], "gt_se_raw": [96.0, 105.0], "gt_se_norm": [0.8016891479492188, 0.970597505569458], "start_token_span": [63, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002164.pt"}
{"idx": 2166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oK6uI4FzYkg.mp4", "duration": 59.934875, "prompt_text": "Please find the visual event described by the sentence 'When does the video list the programs supported by the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video list the programs supported by the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 35.0], "base_se_norm": [-0.19913071393966675, 0.16793429851531982], "gt_se_raw": [26.0, 34.0], "gt_se_norm": [-0.13239163160324097, 0.13456475734710693], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002166.pt"}
{"idx": 2168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oK6uI4FzYkg.mp4", "duration": 59.934875, "prompt_text": "Please find the visual event described by the sentence 'Show me the empty classroom with long tables and chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the empty classroom with long tables and chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [0.4015212059020996, 0.6017385721206665], "gt_se_raw": [42.0, 48.0], "gt_se_norm": [0.4015212059020996, 0.6017385721206665], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002168.pt"}
{"idx": 2170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kvt41MIelDE.mp4", "duration": 105.35524444444445, "prompt_text": "Please find the visual event described by the sentence 'When are fireworks shooting over the Disney castle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are fireworks shooting over the Disney castle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9810166358947754, -0.9050830602645874], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9810166358947754, -0.924066424369812], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002170.pt"}
{"idx": 2172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kvt41MIelDE.mp4", "duration": 105.35524444444445, "prompt_text": "Please find the visual event described by the sentence 'At what time does a man ride a bicycle down a cobblestone street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does a man ride a bicycle down a cobblestone street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.2786310911178589, -0.22168093919754028], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.2786310911178589, -0.22168093919754028], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002172.pt"}
{"idx": 2174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kvt41MIelDE.mp4", "duration": 105.35524444444445, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman floating down from the clouds holding a kite?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman floating down from the clouds holding a kite?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 76.0], "base_se_norm": [0.38578760623931885, 0.4427378177642822], "gt_se_raw": [74.0, 76.0], "gt_se_norm": [0.404771089553833, 0.4427378177642822], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002174.pt"}
{"idx": 2176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJosiEHwWxc.mp4", "duration": 78.91216666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the man in the yellow shirt jumps into the ball pit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the man in the yellow shirt jumps into the ball pit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [0.08981919288635254, 0.16585314273834229], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.11516380310058594, 0.19119775295257568], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002176.pt"}
{"idx": 2178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJosiEHwWxc.mp4", "duration": 78.91216666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the people being told to leave the restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the people being told to leave the restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.3663841485977173, -0.23966103792190552], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [0.368610143661499, 0.44464409351348877], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002178.pt"}
{"idx": 2180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yBiV0NOeXIY.mp4", "duration": 39.6396, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman touching her mouth with her fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman touching her mouth with her fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9495453834533691, -0.7981815934181213], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.8990908265113831], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002180.pt"}
{"idx": 2182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yBiV0NOeXIY.mp4", "duration": 39.6396, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman's vision is simulated as a reduced visual field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman's vision is simulated as a reduced visual field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 32.0], "base_se_norm": [0.2613649368286133, 0.6145471334457397], "gt_se_raw": [25.0, 33.0], "gt_se_norm": [0.2613649368286133, 0.6650017499923706], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002182.pt"}
{"idx": 2184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yBiV0NOeXIY.mp4", "duration": 39.6396, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman brings both hands together with fingers pointing up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman brings both hands together with fingers pointing up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.009091973304748535, 0.16045570373535156], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.8668200969696045, 0.9677293300628662], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002184.pt"}
{"idx": 2186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/usnrNhN_eUA.mp4", "duration": 102.43566666666666, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a man is playing the violin in a white suit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a man is playing the violin in a white suit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.5704621076583862, -0.472839891910553], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.5509376525878906, -0.472839891910553], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002186.pt"}
{"idx": 2188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/usnrNhN_eUA.mp4", "duration": 102.43566666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where people are painting at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where people are painting at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 51.0], "base_se_norm": [-0.08235090970993042, -0.004253089427947998], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.06282645463943481, -0.023777544498443604], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002188.pt"}
{"idx": 2190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-FcsIyqJDzc.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when both the puppy and the adult dog are lying down asleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when both the puppy and the adult dog are lying down asleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, -0.407772958278656], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.5188155174255371], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002190.pt"}
{"idx": 2192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-FcsIyqJDzc.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when a yellow ball is visible on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a yellow ball is visible on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 53.0], "base_se_norm": [-0.0746452808380127, 0.9617520570755005], "gt_se_raw": [25.0, 46.0], "gt_se_norm": [-0.0746452808380127, 0.7026526927947998], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002192.pt"}
{"idx": 2194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-FcsIyqJDzc.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the puppy goes back to snuggle next to the other dog's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the puppy goes back to snuggle next to the other dog's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 53.0], "base_se_norm": [0.5175817012786865, 0.9617520570755005], "gt_se_raw": [40.0, 52.0], "gt_se_norm": [0.4805675745010376, 0.924737811088562], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002194.pt"}
{"idx": 2196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hbOXOHEL0tc.mp4", "duration": 116.58324991658326, "prompt_text": "Please find the visual event described by the sentence 'Find the part where two women are looking at a computer screen together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where two women are looking at a computer screen together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 62.0], "base_se_norm": [0.01215219497680664, 0.06361758708953857], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [-0.005002856254577637, 0.06361758708953857], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002196.pt"}
{"idx": 2198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hbOXOHEL0tc.mp4", "duration": 116.58324991658326, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a woman's feet are shown walking on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a woman's feet are shown walking on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [0.2866342067718506, 0.3724098205566406], "gt_se_raw": [75.0, 80.0], "gt_se_norm": [0.2866342067718506, 0.3724098205566406], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002198.pt"}
{"idx": 2200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6hbd8N3j7bg.mp4", "duration": 113.65520000000001, "prompt_text": "Please find the visual event described by the sentence 'A boy is seen slicing a red apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A boy is seen slicing a red apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.5424758195877075, -0.5072816610336304], "gt_se_raw": [27.0, 28.0], "gt_se_norm": [-0.524878740310669, -0.5072816610336304], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002200.pt"}
{"idx": 2202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6hbd8N3j7bg.mp4", "duration": 113.65520000000001, "prompt_text": "Please find the visual event described by the sentence 'When are the children seen jumping on a trampoline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children seen jumping on a trampoline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 73.0], "base_se_norm": [0.14381039142608643, 0.28458714485168457], "gt_se_raw": [65.0, 73.0], "gt_se_norm": [0.14381039142608643, 0.28458714485168457], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002202.pt"}
{"idx": 2204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6hbd8N3j7bg.mp4", "duration": 113.65520000000001, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the children are getting into the white self-driving car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the children are getting into the white self-driving car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 92.0], "base_se_norm": [0.44296085834503174, 0.6189316511154175], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.49575209617614746, 0.6189316511154175], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002204.pt"}
{"idx": 2206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8rAxqWMKErQ.mp4", "duration": 126.75996666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child attempt to spell a word by tapping on the screen with their fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child attempt to spell a word by tapping on the screen with their fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 60.0], "base_se_norm": [-0.6055536866188049, -0.053328871726989746], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.5424423217773438, -0.4793308973312378], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002206.pt"}
{"idx": 2208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8rAxqWMKErQ.mp4", "duration": 126.75996666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the child is flipping through the pages of a small book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the child is flipping through the pages of a small book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 76.0], "base_se_norm": [-0.005995273590087891, 0.19911682605743408], "gt_se_raw": [65.0, 70.0], "gt_se_norm": [0.025560379028320312, 0.10444962978363037], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002208.pt"}
{"idx": 2210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1VM2eLhvsSM.mp4", "duration": 59.55949999999999, "prompt_text": "Please find the visual event described by the sentence 'When does the camera focus on a man in a white turtleneck sweater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera focus on a man in a white turtleneck sweater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.6306214928627014, -0.4963020086288452], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.5970416069030762, -0.5298818945884705], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002210.pt"}
{"idx": 2212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1VM2eLhvsSM.mp4", "duration": 59.55949999999999, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a man in a cream-colored jacket sings while walking forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a man in a cream-colored jacket sings while walking forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.26124298572540283, -0.12692350149154663], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.2948228716850281, -0.19408327341079712], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002212.pt"}
{"idx": 2214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1VM2eLhvsSM.mp4", "duration": 59.55949999999999, "prompt_text": "Please find the visual event described by the sentence 'When does an aerial shot show the crowd forming a triangular shape on the hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does an aerial shot show the crowd forming a triangular shape on the hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 56.0], "base_se_norm": [0.6118334531784058, 0.8804724216461182], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [0.7461529970169067, 0.9476321935653687], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002214.pt"}
{"idx": 2216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oBDOstVfKQ8.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point is the employment of people with disabilities described as being 'nearly half the rate' of those without?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the employment of people with disabilities described as being 'nearly half the rate' of those without?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9030095934867859, -0.030096113681793213], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.030096113681793213, 0.35786545276641846], "start_token_span": [78, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002216.pt"}
{"idx": 2218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kltuUtE6jQo.mp4", "duration": 50.08, "prompt_text": "Please find the visual event described by the sentence 'When does the '13th CENTURY WOLF' logo fully appear against the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the '13th CENTURY WOLF' logo fully appear against the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 39.0], "base_se_norm": [-0.48083066940307617, 0.5575078725814819], "gt_se_raw": [6.0, 13.0], "gt_se_norm": [-0.7603833675384521, -0.48083066940307617], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002218.pt"}
{"idx": 2220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kltuUtE6jQo.mp4", "duration": 50.08, "prompt_text": "Please find the visual event described by the sentence 'At what time does the laser begin to destroy the logo structure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the laser begin to destroy the logo structure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 39.0], "base_se_norm": [-0.2012779712677002, 0.5575078725814819], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [0.27795517444610596, 0.43769967555999756], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002220.pt"}
{"idx": 2222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kltuUtE6jQo.mp4", "duration": 50.08, "prompt_text": "Please find the visual event described by the sentence 'When does the screen turn to static noise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the screen turn to static noise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 49.0], "base_se_norm": [0.5974440574645996, 0.9568688869476318], "gt_se_raw": [40.0, 49.0], "gt_se_norm": [0.5974440574645996, 0.9568688869476318], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002222.pt"}
{"idx": 2224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9a9XU7D0DIo.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman is doing a yoga pose on a mat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman is doing a yoga pose on a mat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 108.0], "base_se_norm": [0.572724461555481, 0.7332066297531128], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.572724461555481, 0.6369173526763916], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002224.pt"}
{"idx": 2226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9a9XU7D0DIo.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man give a presentation in front of a screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man give a presentation in front of a screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 50.0], "base_se_norm": [-0.35807162523269653, -0.19758951663970947], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.34202343225479126, -0.26178234815597534], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002226.pt"}
{"idx": 2228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mMZ6QRf2Uig.mp4", "duration": 168.16666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of adding the activator to the face mask to form the yellow slime.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of adding the activator to the face mask to form the yellow slime.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 27.0], "base_se_norm": [-0.7978196144104004, -0.6788899898529053], "gt_se_raw": [18.0, 26.0], "gt_se_norm": [-0.7859266996383667, -0.6907829642295837], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002228.pt"}
{"idx": 2230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mMZ6QRf2Uig.mp4", "duration": 168.16666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where red and blue food coloring are dropped into the clear shampoo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where red and blue food coloring are dropped into the clear shampoo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 84.0], "base_se_norm": [-0.060455918312072754, -0.0009911060333251953], "gt_se_raw": [80.0, 84.0], "gt_se_norm": [-0.04856294393539429, -0.0009911060333251953], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002230.pt"}
{"idx": 2232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mMZ6QRf2Uig.mp4", "duration": 168.16666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when sprinkles are poured over the white, edible slime.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 158.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when sprinkles are poured over the white, edible slime.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 158.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [158.0, 163.0], "base_se_norm": [0.8790881633758545, 0.938552975654602], "gt_se_raw": [159.0, 163.0], "gt_se_norm": [0.8909810781478882, 0.938552975654602], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002232.pt"}
{"idx": 2234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YTszmB9fqEs.mp4", "duration": 54.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy say 'my name is Ethan'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy say 'my name is Ethan'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9630086421966553, -0.8520345091819763], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.40813809633255005, -0.3341553807258606], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002234.pt"}
{"idx": 2236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YTszmB9fqEs.mp4", "duration": 54.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the boy and the woman high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the boy and the woman high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 45.0], "base_se_norm": [0.5906288623809814, 0.6646115779876709], "gt_se_raw": [43.0, 44.0], "gt_se_norm": [0.5906288623809814, 0.6276202201843262], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002236.pt"}
{"idx": 2238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0b-EatqUvb4.mp4", "duration": 161.11929166666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black shirt and purple pants dance alone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black shirt and purple pants dance alone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6648446321487427, -0.6276050806045532], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [-0.6524314880371094, -0.6276050806045532], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002238.pt"}
{"idx": 2240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0b-EatqUvb4.mp4", "duration": 161.11929166666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are all the members seen dancing in a circle from a top-down view inside a red-walled cage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are all the members seen dancing in a circle from a top-down view inside a red-walled cage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.39175504446029663, -0.35451555252075195], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.39175504446029663, -0.36692869663238525], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002240.pt"}
{"idx": 2242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0b-EatqUvb4.mp4", "duration": 161.11929166666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man with orange hair is rapping while touching the basketball net.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man with orange hair is rapping while touching the basketball net.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 98.0], "base_se_norm": [0.17925047874450684, 0.21648991107940674], "gt_se_raw": [95.0, 96.0], "gt_se_norm": [0.17925047874450684, 0.19166362285614014], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002242.pt"}
{"idx": 2244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DexH4oCXw-Y.mp4", "duration": 20.487133333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a person is seen sitting in a specialized wheelchair next to a large green object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a person is seen sitting in a specialized wheelchair next to a large green object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9524450302124023], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9524450302124023], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002244.pt"}
{"idx": 2246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DexH4oCXw-Y.mp4", "duration": 20.487133333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a woman says 'sitting really good'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a woman says 'sitting really good'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.023777484893798828, 0.17146694660186768], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.8047555088996887, -0.7071332931518555], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002246.pt"}
{"idx": 2248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tRAPYopBcc.mp4", "duration": 191.39954166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the gray shirt jump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the gray shirt jump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 61.0], "base_se_norm": [-0.4252859354019165, -0.36258989572525024], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.4252859354019165, -0.39393794536590576], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002248.pt"}
{"idx": 2250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tRAPYopBcc.mp4", "duration": 191.39954166666666, "prompt_text": "Please find the visual event described by the sentence 'Where can we see teachers and children sitting in a circle on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see teachers and children sitting in a circle on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.3103431463241577, -0.2580963969230652], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.3103431463241577, -0.2580963969230652], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002250.pt"}
{"idx": 2252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tRAPYopBcc.mp4", "duration": 191.39954166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the shot that shows the exterior of the brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot that shows the exterior of the brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [-0.19540035724639893, -0.1431536078453064], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.19540035724639893, -0.1536029577255249], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002252.pt"}
{"idx": 2254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gu0wmiIngAw.mp4", "duration": 120.45833333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man is lying on a gurney with a woman standing over him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man is lying on a gurney with a woman standing over him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 44.0], "base_se_norm": [-0.3690764307975769, -0.2694569230079651], "gt_se_raw": [39.0, 44.0], "gt_se_norm": [-0.3524731993675232, -0.2694569230079651], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002254.pt"}
{"idx": 2256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gu0wmiIngAw.mp4", "duration": 120.45833333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a woman with glasses speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a woman with glasses speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 63.0], "base_se_norm": [0.012798309326171875, 0.0460047721862793], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [0.012798309326171875, 0.0460047721862793], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002256.pt"}
{"idx": 2258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bR2Uon5BWC0.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the man in the green shirt is repeatedly punching the bald man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the man in the green shirt is repeatedly punching the bald man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 18.0], "base_se_norm": [-0.9631137847900391, -0.7786828279495239], "gt_se_raw": [9.0, 31.0], "gt_se_norm": [-0.889341413974762, -0.6188427209854126], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002258.pt"}
{"idx": 2260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bR2Uon5BWC0.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the men are trying to pull the bald man off of the other man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the men are trying to pull the bald man off of the other man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 68.0], "base_se_norm": [-0.42211633920669556, -0.1639130711555481], "gt_se_raw": [47.0, 68.0], "gt_se_norm": [-0.42211633920669556, -0.1639130711555481], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002260.pt"}
{"idx": 2262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bR2Uon5BWC0.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the blue denim jacket cover the injured man's hand with a cloth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the blue denim jacket cover the injured man's hand with a cloth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 103.0], "base_se_norm": [-0.05325448513031006, 0.26642584800720215], "gt_se_raw": [89.0, 93.0], "gt_se_norm": [0.09429025650024414, 0.1434718370437622], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002262.pt"}
{"idx": 2264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdQwlOy78lg.mp4", "duration": 101.1, "prompt_text": "Please find the visual event described by the sentence 'Please provide the time range when the person first writes a signature on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please provide the time range when the person first writes a signature on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.46587538719177246, -0.386745810508728], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.46587538719177246, -0.40652817487716675], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002264.pt"}
{"idx": 2266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdQwlOy78lg.mp4", "duration": 101.1, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the person writes 'Andrew' on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the person writes 'Andrew' on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 71.0], "base_se_norm": [0.167161226272583, 0.4045499563217163], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.167161226272583, 0.24629080295562744], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002266.pt"}
{"idx": 2268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4w7sVSMbjyM.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the snowboarders cross the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the snowboarders cross the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 84.0], "base_se_norm": [-0.16808420419692993, -0.11543130874633789], "gt_se_raw": [79.0, 84.0], "gt_se_norm": [-0.16808420419692993, -0.11543130874633789], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002268.pt"}
{"idx": 2270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4w7sVSMbjyM.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when one snowboarder falls while the other passes by.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when one snowboarder falls while the other passes by.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 75.0], "base_se_norm": [-0.2417982816696167, -0.2102065086364746], "gt_se_raw": [162.0, 165.0], "gt_se_norm": [0.7059539556503296, 0.7375457286834717], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002270.pt"}
{"idx": 2272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4w7sVSMbjyM.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the different 'skootch' moves demonstrated on a white background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 174.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the different 'skootch' moves demonstrated on a white background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 174.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 174.0], "base_se_norm": [0.7586067914962769, 0.8323209285736084], "gt_se_raw": [167.0, 174.0], "gt_se_norm": [0.7586067914962769, 0.8323209285736084], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002272.pt"}
{"idx": 2274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w38j0ea8HO4.mp4", "duration": 126.86006666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with white skin state that he is an android, not a robot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with white skin state that he is an android, not a robot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [-0.2905569076538086, -0.1486682891845703], "gt_se_raw": [100.0, 102.0], "gt_se_norm": [0.5765402317047119, 0.608070969581604], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002274.pt"}
{"idx": 2276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w38j0ea8HO4.mp4", "duration": 126.86006666666667, "prompt_text": "Please find the visual event described by the sentence 'When are two people serving ten chocolate sundaes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two people serving ten chocolate sundaes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 126.0], "base_se_norm": [0.8760828971862793, 0.9864406585693359], "gt_se_raw": [120.0, 123.0], "gt_se_norm": [0.8918482065200806, 0.939144492149353], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002276.pt"}
{"idx": 2278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggqrEMtA3TU.mp4", "duration": 377.28, "prompt_text": "Please find the visual event described by the sentence 'At what point do the two men shake hands for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the two men shake hands for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 116.0], "base_se_norm": [-0.40627652406692505, -0.38507211208343506], "gt_se_raw": [112.0, 116.0], "gt_se_norm": [-0.40627652406692505, -0.38507211208343506], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002278.pt"}
{"idx": 2280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggqrEMtA3TU.mp4", "duration": 377.28, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the two men switch their seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 237.0 - 242.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the two men switch their seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 237.0 - 242.0 seconds.<|im_end|>\n", "base_se_raw": [237.0, 242.0], "base_se_norm": [0.25636136531829834, 0.28286683559417725], "gt_se_raw": [237.0, 242.0], "gt_se_norm": [0.25636136531829834, 0.28286683559417725], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002280.pt"}
{"idx": 2282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggqrEMtA3TU.mp4", "duration": 377.28, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where one man hands a piece of paper to the other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 314.0 - 327.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where one man hands a piece of paper to the other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 314.0 - 327.0 seconds.<|im_end|>\n", "base_se_raw": [314.0, 327.0], "base_se_norm": [0.6645462512969971, 0.733460545539856], "gt_se_raw": [321.0, 323.0], "gt_se_norm": [0.7016539573669434, 0.7122561931610107], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002282.pt"}
{"idx": 2284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tw7PR89IOI.mp4", "duration": 40.073366666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the boy is closing the mailbox door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the boy is closing the mailbox door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [0.6968876123428345, 0.8965214490890503], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [0.6968876123428345, 0.8466129302978516], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002284.pt"}
{"idx": 2286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tw7PR89IOI.mp4", "duration": 40.073366666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman helps to position the wheelchair in front of the mailboxes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman helps to position the wheelchair in front of the mailboxes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 26.0], "base_se_norm": [-0.15155619382858276, 0.29761993885040283], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [0.04807758331298828, 0.347528338432312], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002286.pt"}
{"idx": 2288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGAVcg_xUJE.mp4", "duration": 131.56476666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man pointing his right index finger while holding a cigar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man pointing his right index finger while holding a cigar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.8479835987091064, -0.8175803422927856], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.832781970500946, -0.8023787140846252], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002288.pt"}
{"idx": 2290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGAVcg_xUJE.mp4", "duration": 131.56476666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the text 'most successful defensive organization...' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the text 'most successful defensive organization...' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.43753939867019653, -0.34632956981658936], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.43753939867019653, -0.37673288583755493], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002290.pt"}
{"idx": 2292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGAVcg_xUJE.mp4", "duration": 131.56476666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man bring the cigar to his mouth and yell 'Hoo-yah'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man bring the cigar to his mouth and yell 'Hoo-yah'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [-0.6047574281692505, -0.5591524839401245], "gt_se_raw": [129.0, 131.0], "gt_se_norm": [0.9610112905502319, 0.9914145469665527], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002292.pt"}
{"idx": 2294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3OhB-y7Tls8.mp4", "duration": 132.2146064760104, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the three main points: Positive, Problematic, and Possible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the three main points: Positive, Problematic, and Possible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 19.0], "base_se_norm": [-0.8336038589477539, -0.7125884294509888], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.8336038589477539, -0.7579692602157593], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002294.pt"}
{"idx": 2296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3OhB-y7Tls8.mp4", "duration": 132.2146064760104, "prompt_text": "Please find the visual event described by the sentence 'When does the video show people with a rickshaw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show people with a rickshaw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 101.0], "base_se_norm": [0.4673115015029907, 0.5278191566467285], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [0.4673115015029907, 0.5126922130584717], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002296.pt"}
{"idx": 2298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HvPbH30KWLE.mp4", "duration": 119.61949999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we observe someone eating a piece of fruit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe someone eating a piece of fruit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.9331212639808655, -0.8996819257736206], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9331212639808655, -0.8996819257736206], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002298.pt"}
{"idx": 2300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HvPbH30KWLE.mp4", "duration": 119.61949999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the group is lined up outdoors in front of a large green hedge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the group is lined up outdoors in front of a large green hedge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 58.0], "base_se_norm": [-0.14729619026184082, -0.0302584171295166], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.14729619026184082, -0.06369781494140625], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002300.pt"}
{"idx": 2302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HvPbH30KWLE.mp4", "duration": 119.61949999999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when two men are playing a dancing arcade game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when two men are playing a dancing arcade game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 74.0], "base_se_norm": [0.187097430229187, 0.23725652694702148], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [0.20381712913513184, 0.23725652694702148], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002302.pt"}
{"idx": 2304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQD6gJMnWVs.mp4", "duration": 112.71260000000001, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a person walking up an escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a person walking up an escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.48541688919067383, -0.3257187008857727], "gt_se_raw": [29.0, 35.0], "gt_se_norm": [-0.48541688919067383, -0.37895143032073975], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002304.pt"}
{"idx": 2306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQD6gJMnWVs.mp4", "duration": 112.71260000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a red car pulls into a parking spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a red car pulls into a parking spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [0.011421918869018555, 0.06465470790863037], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [0.029166221618652344, 0.06465470790863037], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002306.pt"}
{"idx": 2308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yqFxYu15J6E.mp4", "duration": 35.618916666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the woman use hand-over-hand guidance to communicate with the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman use hand-over-hand guidance to communicate with the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.943850040435791, -0.5508004426956177], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.943850040435791, -0.5508004426956177], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002308.pt"}
{"idx": 2310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yqFxYu15J6E.mp4", "duration": 35.618916666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when the girl is touching and reading the calendar with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when the girl is touching and reading the calendar with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.2700507640838623, 0.12299883365631104], "gt_se_raw": [13.0, 19.0], "gt_se_norm": [-0.2700507640838623, 0.06684887409210205], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002310.pt"}
{"idx": 2312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yqFxYu15J6E.mp4", "duration": 35.618916666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the girl places an orange sticker on the calendar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the girl places an orange sticker on the calendar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.2700507640838623, 0.12299883365631104], "gt_se_raw": [27.0, 33.0], "gt_se_norm": [0.5160484313964844, 0.8529480695724487], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002312.pt"}
{"idx": 2314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CecE0asmNwc.mp4", "duration": 188.39654166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a person puts a wheelchair into a cleaning machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a person puts a wheelchair into a cleaning machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 38.0], "base_se_norm": [-0.6602910161018372, -0.5965955853462219], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.6602910161018372, -0.6072114706039429], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002314.pt"}
{"idx": 2316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CecE0asmNwc.mp4", "duration": 188.39654166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a worker is cleaning the wheels of a wheelchair with a high-pressure spray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a worker is cleaning the wheels of a wheelchair with a high-pressure spray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.6921387314796448, -0.6496751308441162], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.6921387314796448, -0.6602910161018372], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002316.pt"}
{"idx": 2318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CeN6PMEOQEA.mp4", "duration": 147.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man pour the red drink into the glass of water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man pour the red drink into the glass of water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 63.0], "base_se_norm": [-0.2521246075630188, -0.1433427333831787], "gt_se_raw": [56.0, 62.0], "gt_se_norm": [-0.23852688074111938, -0.15694046020507812], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002318.pt"}
{"idx": 2320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CeN6PMEOQEA.mp4", "duration": 147.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man stir the drink with a yellow stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man stir the drink with a yellow stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 91.0], "base_se_norm": [0.14220964908599854, 0.23739385604858398], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.15580737590789795, 0.2509915828704834], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002320.pt"}
{"idx": 2322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CeN6PMEOQEA.mp4", "duration": 147.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the 'BADLANDS CHUGS' logo appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the 'BADLANDS CHUGS' logo appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.9728045463562012, -0.9184135794639587], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.9592068195343018, -0.9184135794639587], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002322.pt"}
{"idx": 2324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u3VVUu-lZsM.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention the employment rate for people without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention the employment rate for people without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9030095934867859, -0.7090288400650024], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002324.pt"}
{"idx": 2326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u3VVUu-lZsM.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator state that disabled people are employed at 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator state that disabled people are employed at 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.030096113681793213, 0.35786545276641846], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002326.pt"}
{"idx": 2328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eA-GJNebJj0.mp4", "duration": 69.70296666666667, "prompt_text": "Please find the visual event described by the sentence 'A hand touches a small leaf on a branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A hand touches a small leaf on a branch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8278409242630005, -0.684374988079071], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.8278409242630005, -0.741761326789856], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002328.pt"}
{"idx": 2330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eA-GJNebJj0.mp4", "duration": 69.70296666666667, "prompt_text": "Please find the visual event described by the sentence 'A person walks through the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person walks through the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 38.0], "base_se_norm": [-0.1392045021057129, 0.09034097194671631], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.11051130294799805, 0.004261493682861328], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002330.pt"}
{"idx": 2332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eA-GJNebJj0.mp4", "duration": 69.70296666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the old man's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the old man's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [0.4633524417877197, 0.6068183183670044], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [0.492045521736145, 0.6068183183670044], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002332.pt"}
{"idx": 2334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6nX0lYo0Uno.mp4", "duration": 98.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where he mentions being an ICT major.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where he mentions being an ICT major.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 22.0], "base_se_norm": [-0.6559865474700928, -0.554806113243103], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.8178752064704895, -0.7774030566215515], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002334.pt"}
{"idx": 2336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6nX0lYo0Uno.mp4", "duration": 98.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does he express his desire to stay in Lexington and find work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does he express his desire to stay in Lexington and find work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.595278263092041, -0.3929173946380615], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.2917369604110718, -0.23102867603302002], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002336.pt"}
{"idx": 2338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGMab_ZhQTM.mp4", "duration": 134.63040304234153, "prompt_text": "Please find the visual event described by the sentence 'When is the man seen from a high-angle, over-the-shoulder shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man seen from a high-angle, over-the-shoulder shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.8514451384544373], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.925722599029541], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002338.pt"}
{"idx": 2340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGMab_ZhQTM.mp4", "duration": 134.63040304234153, "prompt_text": "Please find the visual event described by the sentence 'Show me the time range where the man talks about the heavy, old-style 'Perkins Brailler'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time range where the man talks about the heavy, old-style 'Perkins Brailler'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 52.0], "base_se_norm": [-0.8514451384544373, -0.22751474380493164], "gt_se_raw": [36.0, 55.0], "gt_se_norm": [-0.4652025103569031, -0.1829482913017273], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002340.pt"}
{"idx": 2342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGMab_ZhQTM.mp4", "duration": 134.63040304234153, "prompt_text": "Please find the visual event described by the sentence 'When does the man identify his current device as a 'BrailleNote' and describe its evolution and modern features?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man identify his current device as a 'BrailleNote' and describe its evolution and modern features?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 30.0], "base_se_norm": [-0.8514451384544373, -0.5543354749679565], "gt_se_raw": [117.0, 134.0], "gt_se_norm": [0.7380918264389038, 0.9906350374221802], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002342.pt"}
{"idx": 2344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1WsDtn-feuI.mp4", "duration": 13.066666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man is flung to the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man is flung to the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.3877550959587097, 0.22448980808258057], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.5408163070678711, -0.23469388484954834], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002344.pt"}
{"idx": 2346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1WsDtn-feuI.mp4", "duration": 13.066666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the dog eat the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog eat the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.5306122303009033, 0.8367346525192261], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [0.5306122303009033, 0.6836735010147095], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002346.pt"}
{"idx": 2348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NAizsUKpafs.mp4", "duration": 129.5294, "prompt_text": "Please find the visual event described by the sentence 'When does the woman say that her son absolutely refuses to get back in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman say that her son absolutely refuses to get back in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 122.0], "base_se_norm": [0.8374205827713013, 0.8837422132492065], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [-0.4441416263580322, -0.41326063871383667], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002348.pt"}
{"idx": 2350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NAizsUKpafs.mp4", "duration": 129.5294, "prompt_text": "Please find the visual event described by the sentence 'Show the photograph of the family at the Pikes Peak summit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the photograph of the family at the Pikes Peak summit.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 102.0], "base_se_norm": [0.49772942066192627, 0.5749319791793823], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.5131698846817017, 0.5749319791793823], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002350.pt"}
{"idx": 2352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NAizsUKpafs.mp4", "duration": 129.5294, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the speaker uses the dinner routine as an example of providing structure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the speaker uses the dinner routine as an example of providing structure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 68.0], "base_se_norm": [-0.16621249914169312, 0.04995465278625488], "gt_se_raw": [111.0, 119.0], "gt_se_norm": [0.7138965129852295, 0.8374205827713013], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002352.pt"}
{"idx": 2354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTTxroRR-G0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does an instructor address a group of adults while sitting on a green beanbag chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does an instructor address a group of adults while sitting on a green beanbag chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002354.pt"}
{"idx": 2356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTTxroRR-G0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a toddler is laughing next to a smiling woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a toddler is laughing next to a smiling woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002356.pt"}
{"idx": 2358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/huT5__BqY_U.mp4", "duration": 166.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the cows are crowded in the pen, waiting to be let out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the cows are crowded in the pen, waiting to be let out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.7236683964729309], "gt_se_raw": [0.0, 23.0], "gt_se_norm": [-1.0, -0.7236683964729309], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002358.pt"}
{"idx": 2360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/huT5__BqY_U.mp4", "duration": 166.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the cows running and jumping joyfully in the grassy field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the cows running and jumping joyfully in the grassy field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 80.0], "base_se_norm": [-0.5554665327072144, -0.03884655237197876], "gt_se_raw": [38.0, 75.0], "gt_se_norm": [-0.5434521436691284, -0.09891867637634277], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002360.pt"}
{"idx": 2362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/huT5__BqY_U.mp4", "duration": 166.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when a cow with the number '13' on its flank is clearly visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a cow with the number '13' on its flank is clearly visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 132.0], "base_se_norm": [0.5138165950775146, 0.5859031677246094], "gt_se_raw": [126.0, 131.0], "gt_se_norm": [0.5138165950775146, 0.5738887786865234], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002362.pt"}
{"idx": 2364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQ7YxcW9Vyk.mp4", "duration": 91.82506666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where a young man is adjusting his red helmet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where a young man is adjusting his red helmet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [0.6771019697189331, 0.8077852725982666], "gt_se_raw": [77.0, 84.0], "gt_se_norm": [0.6771019697189331, 0.8295657634735107], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002364.pt"}
{"idx": 2366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQ7YxcW9Vyk.mp4", "duration": 91.82506666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman smile while looking at a baby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman smile while looking at a baby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 41.0], "base_se_norm": [-0.303022563457489, -0.10699766874313354], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.303022563457489, -0.10699766874313354], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002366.pt"}
{"idx": 2368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s_RV4Btuv2c.mp4", "duration": 36.3363, "prompt_text": "Please find the visual event described by the sentence 'A man is climbing a rock wall with the help of a woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is climbing a rock wall with the help of a woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 14.0], "base_se_norm": [-0.9449586272239685, -0.2294207215309143], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9449586272239685, -0.1743793487548828], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002368.pt"}
{"idx": 2370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s_RV4Btuv2c.mp4", "duration": 36.3363, "prompt_text": "Please find the visual event described by the sentence 'A person guides the man's hand to place the pot on the stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person guides the man's hand to place the pot on the stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.37603437900543213, 0.5411585569381714], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [0.37603437900543213, 0.5411585569381714], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002370.pt"}
{"idx": 2372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s_RV4Btuv2c.mp4", "duration": 36.3363, "prompt_text": "Please find the visual event described by the sentence 'A bus driver is working while it's snowing outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A bus driver is working while it's snowing outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [0.5961999893188477, 0.7613240480422974], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [0.5961999893188477, 0.7062827348709106], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002372.pt"}
{"idx": 2374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SBTL1vI4-mc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person is manually drawing a blueprint.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person is manually drawing a blueprint.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47478586435317993, -0.08087527751922607], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002374.pt"}
{"idx": 2376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SBTL1vI4-mc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a red shirt seen working at a drafting table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a red shirt seen working at a drafting table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4443387985229492, 0.8382494449615479], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4443387985229492, 0.8382494449615479], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002376.pt"}
{"idx": 2378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YvcTBNSuiGk.mp4", "duration": 231.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the black shirt ask if she is a farm animal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the black shirt ask if she is a farm animal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.9050769209861755, -0.8619301319122314], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.5857902765274048, -0.5685315728187561], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002378.pt"}
{"idx": 2380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YvcTBNSuiGk.mp4", "duration": 231.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy use the pen to cross something off the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy use the pen to cross something off the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 35.0], "base_se_norm": [-0.767007052898407, -0.6979720592498779], "gt_se_raw": [28.0, 34.0], "gt_se_norm": [-0.7583776712417603, -0.7066015005111694], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002380.pt"}
{"idx": 2382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YvcTBNSuiGk.mp4", "duration": 231.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman in black asks if she is a pet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman in black asks if she is a pet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [-0.689342737197876, -0.6289371252059937], "gt_se_raw": [130.0, 132.0], "gt_se_norm": [0.121817946434021, 0.13907670974731445], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002382.pt"}
{"idx": 2384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_NWYeVyZz9I.mp4", "duration": 176.09258333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the moose character add purple meatballs to the stew?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the moose character add purple meatballs to the stew?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [-0.6819854974746704, -0.6479125022888184], "gt_se_raw": [30.0, 31.0], "gt_se_norm": [-0.659270167350769, -0.6479125022888184], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002384.pt"}
{"idx": 2386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_NWYeVyZz9I.mp4", "duration": 176.09258333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the man in the blue polo shirt adding purple lizards to the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the man in the blue polo shirt adding purple lizards to the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 81.0], "base_se_norm": [-0.18224841356277466, -0.0800294280052185], "gt_se_raw": [81.0, 82.0], "gt_se_norm": [-0.0800294280052185, -0.06867176294326782], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002386.pt"}
{"idx": 2388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fSVp2m1ahwk.mp4", "duration": 29.56, "prompt_text": "Please find the visual event described by the sentence 'When is a woman in a black blazer speaking at a podium with the University of Minnesota logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman in a black blazer speaking at a podium with the University of Minnesota logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.7970229983329773, -0.6617050170898438], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.7970229983329773, -0.6617050170898438], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002388.pt"}
{"idx": 2390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fSVp2m1ahwk.mp4", "duration": 29.56, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman appear, singing into a vintage microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman appear, singing into a vintage microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.3234099745750427, -0.18809199333190918], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [-0.3910689949989319, -0.3234099745750427], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002390.pt"}
{"idx": 2392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fSVp2m1ahwk.mp4", "duration": 29.56, "prompt_text": "Please find the visual event described by the sentence 'Find the shot of a large group posing for a photo outside a brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot of a large group posing for a photo outside a brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 24.0], "base_se_norm": [0.35318005084991455, 0.6238160133361816], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [0.42083895206451416, 0.6238160133361816], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002392.pt"}
{"idx": 2394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-8Th1fNuXA.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment two men are seen examining a yellow robotic arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment two men are seen examining a yellow robotic arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002394.pt"}
{"idx": 2396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-8Th1fNuXA.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman smiling in front of a robotic arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman smiling in front of a robotic arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002396.pt"}
{"idx": 2398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsEOupmYmF0.mp4", "duration": 156.22273333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the woman seen eating in an indoor cafeteria setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman seen eating in an indoor cafeteria setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.9103843569755554, -0.8079664707183838], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.9103843569755554, -0.8079664707183838], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002398.pt"}
{"idx": 2400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsEOupmYmF0.mp4", "duration": 156.22273333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman is walking down the street while eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman is walking down the street while eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.37269049882888794, -0.3214815855026245], "gt_se_raw": [48.0, 53.0], "gt_se_norm": [-0.385492742061615, -0.3214815855026245], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002400.pt"}
{"idx": 2402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsEOupmYmF0.mp4", "duration": 156.22273333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when the young girl starts eating the olives.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when the young girl starts eating the olives.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.9103843569755554, -0.8079664707183838], "gt_se_raw": [143.0, 151.0], "gt_se_norm": [0.8307194709777832, 0.9331372976303101], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002402.pt"}
{"idx": 2404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cEfD-Mr7m2c.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the part of the video that displays the 79.1% employment rate for non-disabled Americans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part of the video that displays the 79.1% employment rate for non-disabled Americans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [6.0, 11.0], "gt_se_norm": [-0.4180576801300049, 0.0668942928314209], "start_token_span": [82, 85], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002404.pt"}
{"idx": 2406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cEfD-Mr7m2c.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the data point for the employment of disabled individuals in 2010.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the data point for the employment of disabled individuals in 2010.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.6120384335517883, -0.12708652019500732], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002406.pt"}
{"idx": 2408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nqkyzpaoMug.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a student is combing a mannequin's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a student is combing a mannequin's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.8032602071762085, -0.5409404635429382], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.7376802563667297, -0.5409404635429382], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002408.pt"}
{"idx": 2410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nqkyzpaoMug.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a mannequin head is decorated with orange spiders.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a mannequin head is decorated with orange spiders.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8032602071762085], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002410.pt"}
{"idx": 2412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nqkyzpaoMug.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Can you show me when a woman with blonde hair is smiling at the camera with her arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you show me when a woman with blonde hair is smiling at the camera with her arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002412.pt"}
{"idx": 2414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KTtkpl7T9sg.mp4", "duration": 118.70191666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time the woman anxiously looks around while holding her phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time the woman anxiously looks around while holding her phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 83.0], "base_se_norm": [-0.8315107226371765, 0.3984609842300415], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.37658971548080444, -0.27549612522125244], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002414.pt"}
{"idx": 2416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KTtkpl7T9sg.mp4", "duration": 118.70191666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman crosses her arms after her phone dies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman crosses her arms after her phone dies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 92.0], "base_se_norm": [0.38161206245422363, 0.5501012802124023], "gt_se_raw": [88.0, 91.0], "gt_se_norm": [0.48270559310913086, 0.5332523584365845], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002416.pt"}
{"idx": 2418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUenwQAtW8M.mp4", "duration": 84.65123333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the black dog appear at the bottom of the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the black dog appear at the bottom of the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 52.0], "base_se_norm": [-0.1021985411643982, 0.22857046127319336], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.1021985411643982, -0.03131943941116333], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002418.pt"}
{"idx": 2420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUenwQAtW8M.mp4", "duration": 84.65123333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the black dog walk up the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the black dog walk up the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 63.0], "base_se_norm": [0.13406503200531006, 0.4884603023529053], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [0.22857046127319336, 0.3467022180557251], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002420.pt"}
{"idx": 2422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUenwQAtW8M.mp4", "duration": 84.65123333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the golden retriever finally runs down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the golden retriever finally runs down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 74.0], "base_se_norm": [0.6774711608886719, 0.7483502626419067], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [0.6774711608886719, 0.7247239351272583], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002422.pt"}
{"idx": 2424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-mJqm_ijE6s.mp4", "duration": 82.61586666666666, "prompt_text": "Please find the visual event described by the sentence 'When does a woman hand out gifts to everyone at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman hand out gifts to everyone at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 43.0], "base_se_norm": [-0.3947893977165222, 0.04096221923828125], "gt_se_raw": [29.0, 34.0], "gt_se_norm": [-0.2979556918144226, -0.1769135594367981], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002424.pt"}
{"idx": 2426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-mJqm_ijE6s.mp4", "duration": 82.61586666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man proposes a toast to Harry Potter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man proposes a toast to Harry Potter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 77.0], "base_se_norm": [0.7914234399795532, 0.8640487194061279], "gt_se_raw": [62.0, 69.0], "gt_se_norm": [0.5009223222732544, 0.6703813076019287], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002426.pt"}
{"idx": 2428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rCZ3SN65kIs.mp4", "duration": 272.98104166666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the Witch-king is seen flying towards the city on his beast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the Witch-king is seen flying towards the city on his beast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.9047552943229675, -0.8827757239341736], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.9047552943229675, -0.8827757239341736], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002428.pt"}
{"idx": 2430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rCZ3SN65kIs.mp4", "duration": 272.98104166666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where Th\u00e9oden shouts 'Arise, Riders of Th\u00e9oden!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where Th\u00e9oden shouts 'Arise, Riders of Th\u00e9oden!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 158.0], "base_se_norm": [0.13560998439788818, 0.15758955478668213], "gt_se_raw": [99.0, 102.0], "gt_se_norm": [-0.2746749520301819, -0.25269538164138794], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002430.pt"}
{"idx": 2432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rCZ3SN65kIs.mp4", "duration": 272.98104166666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the cavalry charge first makes impact with the enemy lines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 208.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the cavalry charge first makes impact with the enemy lines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 208.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 208.0], "base_se_norm": [0.4066910743713379, 0.5239152908325195], "gt_se_raw": [243.0, 246.0], "gt_se_norm": [0.7803434133529663, 0.8023228645324707], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002432.pt"}
{"idx": 2434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aHi8idsTvVM.mp4", "duration": 96.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Miranda takes an order from the woman with blonde hair and glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Miranda takes an order from the woman with blonde hair and glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 33.0], "base_se_norm": [-0.8554216623306274, -0.3184165358543396], "gt_se_raw": [7.0, 24.0], "gt_se_norm": [-0.8554216623306274, -0.504302978515625], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002434.pt"}
{"idx": 2436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aHi8idsTvVM.mp4", "duration": 96.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the finished drinks being put into a carrier for transport?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the finished drinks being put into a carrier for transport?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 68.0], "base_se_norm": [0.21858859062194824, 0.40447497367858887], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [0.3012048006057739, 0.40447497367858887], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002436.pt"}
{"idx": 2438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oKficmlxzaI.mp4", "duration": 190.4903, "prompt_text": "Please find the visual event described by the sentence 'When can we see two eagles sitting on a tree branch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two eagles sitting on a tree branch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.7480186223983765, -0.7165210247039795], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7480186223983765, -0.7165210247039795], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002438.pt"}
{"idx": 2440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oKficmlxzaI.mp4", "duration": 190.4903, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the Great Seal of the USA is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the Great Seal of the USA is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.4960373044013977, -0.4435412287712097], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.4960373044013977, -0.45404040813446045], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002440.pt"}
{"idx": 2442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oKficmlxzaI.mp4", "duration": 190.4903, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when an eagle is eating the fish it caught while perched on a tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 153.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when an eagle is eating the fish it caught while perched on a tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 153.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [153.0, 162.0], "base_se_norm": [0.6063810586929321, 0.7008740901947021], "gt_se_raw": [154.0, 162.0], "gt_se_norm": [0.6168802976608276, 0.7008740901947021], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002442.pt"}
{"idx": 2444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Grz6lB5YX9I.mp4", "duration": 22.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman tickles the baby's chin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman tickles the baby's chin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.6390976905822754, -0.27819544076919556], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.45864659547805786, -0.27819544076919556], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002444.pt"}
{"idx": 2446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Grz6lB5YX9I.mp4", "duration": 22.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the baby makes a frowning face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the baby makes a frowning face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.4436091184616089, 0.6240601539611816], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [0.7142857313156128, 0.8947368860244751], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002446.pt"}
{"idx": 2448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JOCtdw9FG-s.mp4", "duration": 83.79204166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the chef chop the shallots?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the chef chop the shallots?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.7851824760437012, -0.7374452352523804], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.880656898021698, -0.8567882776260376], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002448.pt"}
{"idx": 2450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JOCtdw9FG-s.mp4", "duration": 83.79204166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the tiny dining table being set up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the tiny dining table being set up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.45102185010910034, -0.35554736852645874], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.45102185010910034, -0.35554736852645874], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002450.pt"}
{"idx": 2452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JOCtdw9FG-s.mp4", "duration": 83.79204166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the hamster eating the burrito from a wide-angle shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the hamster eating the burrito from a wide-angle shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 56.0], "base_se_norm": [-0.09299260377883911, 0.3366425037384033], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.11686122417449951, 0.0024819374084472656], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002452.pt"}
{"idx": 2454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Osw1_0jMM4M.mp4", "duration": 127.36056666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the little girl in a red traditional dress with umbrellas in her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the little girl in a red traditional dress with umbrellas in her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 37.0], "base_se_norm": [-0.623117208480835, -0.4189724326133728], "gt_se_raw": [24.0, 37.0], "gt_se_norm": [-0.623117208480835, -0.4189724326133728], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002454.pt"}
{"idx": 2456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Osw1_0jMM4M.mp4", "duration": 127.36056666666667, "prompt_text": "Please find the visual event described by the sentence 'Please locate the scene with two fishermen on bamboo rafts at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please locate the scene with two fishermen on bamboo rafts at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 87.0], "base_se_norm": [0.3190896511077881, 0.36619997024536133], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [0.3190896511077881, 0.36619997024536133], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002456.pt"}
{"idx": 2458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzWq3LqxAIc.mp4", "duration": 23.857166666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see the person scooping cat food from the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the person scooping cat food from the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.49700647592544556, -0.07784521579742432], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.41317421197891235, -0.07784521579742432], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002458.pt"}
{"idx": 2460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzWq3LqxAIc.mp4", "duration": 23.857166666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person is opening the food container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person is opening the food container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.8323354721069336, -0.5808387398719788], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.8323354721069336, -0.5808387398719788], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002460.pt"}
{"idx": 2462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzWq3LqxAIc.mp4", "duration": 23.857166666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the scoop is placed back into the container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the scoop is placed back into the container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.3413161039352417, 0.5089805126190186], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [0.2574838399887085, 0.42514824867248535], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002462.pt"}
{"idx": 2464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S0zfR9DTwSY.mp4", "duration": 15.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the teacher in the grey shirt adjusts her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the teacher in the grey shirt adjusts her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.23890060186386108, 0.014799118041992188], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.23890060186386108, 0.014799118041992188], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002464.pt"}
{"idx": 2466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S0zfR9DTwSY.mp4", "duration": 15.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the boy in the white tank top on the left is shouting excitedly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the boy in the white tank top on the left is shouting excitedly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [0.1416490077972412, 0.3953489065170288], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.6490486860275269, 0.9027484655380249], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002466.pt"}
{"idx": 2468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3cOSt2QVJf4.mp4", "duration": 123.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a cow wearing sunglasses and playing the violin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a cow wearing sunglasses and playing the violin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 27.0], "base_se_norm": [-0.7731568813323975, -0.5625168681144714], "gt_se_raw": [15.0, 27.0], "gt_se_norm": [-0.7569538354873657, -0.5625168681144714], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002468.pt"}
{"idx": 2470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3cOSt2QVJf4.mp4", "duration": 123.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do the animals promenade two by two?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the animals promenade two by two?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 62.0], "base_se_norm": [-0.1574399471282959, 0.004590868949890137], "gt_se_raw": [52.0, 62.0], "gt_se_norm": [-0.1574399471282959, 0.004590868949890137], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002470.pt"}
{"idx": 2472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3cOSt2QVJf4.mp4", "duration": 123.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the barnyard dog is spinning with another animal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the barnyard dog is spinning with another animal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 99.0], "base_se_norm": [0.47448015213012695, 0.60410475730896], "gt_se_raw": [91.0, 99.0], "gt_se_norm": [0.47448015213012695, 0.60410475730896], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002472.pt"}
{"idx": 2474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymPpIzaanhY.mp4", "duration": 5.633333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man says 'You lose'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man says 'You lose'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [0.06508874893188477, 0.7751479148864746], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.28994083404541016, 0.4201183319091797], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002474.pt"}
{"idx": 2476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymPpIzaanhY.mp4", "duration": 5.633333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the man says 'Good day, sir'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the man says 'Good day, sir'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.28994083404541016], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [0.4201183319091797, 0.7751479148864746], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002476.pt"}
{"idx": 2478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qoh3YkxuwVo.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the black jacket is walking down the long corridor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the black jacket is walking down the long corridor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 13.0], "base_se_norm": [-0.9381423592567444, -0.8391700983047485], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.9133992791175842, -0.8391700983047485], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002478.pt"}
{"idx": 2480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qoh3YkxuwVo.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the older man says, 'You don't have to prove nothin' to nobody except yourself.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the older man says, 'You don't have to prove nothin' to nobody except yourself.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 97.0], "base_se_norm": [0.08869457244873047, 0.20003831386566162], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.19585061073303223, -0.158735990524292], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002480.pt"}
{"idx": 2482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qoh3YkxuwVo.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the older man reveals his regret by saying, 'Still not a week goes by I don't regret it.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the older man reveals his regret by saying, 'Still not a week goes by I don't regret it.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 97.0], "base_se_norm": [0.11343765258789062, 0.20003831386566162], "gt_se_raw": [113.0, 116.0], "gt_se_norm": [0.3979828357696533, 0.43509745597839355], "start_token_span": [81, 85], "end_token_span": [87, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002482.pt"}
{"idx": 2484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vcrOYQbg3hY.mp4", "duration": 115.40695833333332, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the nameplate on the office door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the nameplate on the office door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 29.0], "base_se_norm": [-0.6014105081558228, -0.497430682182312], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.5840805768966675, -0.5320906043052673], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002484.pt"}
{"idx": 2486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vcrOYQbg3hY.mp4", "duration": 115.40695833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the text 'What are the challenges for your division?' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the text 'What are the challenges for your division?' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 54.0], "base_se_norm": [-0.09884119033813477, -0.06418120861053467], "gt_se_raw": [52.0, 53.0], "gt_se_norm": [-0.09884119033813477, -0.08151119947433472], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002486.pt"}
{"idx": 2488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oiX7oBoavlA.mp4", "duration": 154.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'When do two identical characters fight each other on a green island floating in space?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two identical characters fight each other on a green island floating in space?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9481417536735535], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9611063003540039], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002488.pt"}
{"idx": 2490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oiX7oBoavlA.mp4", "duration": 154.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does a character with a red cape kick another character wearing blue, causing a large pink explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a character with a red cape kick another character wearing blue, causing a large pink explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.5073465704917908, -0.4684528708457947], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.10544508695602417, -0.07951599359512329], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002490.pt"}
{"idx": 2492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TGrF7g4Z8EE.mp4", "duration": 135.59379166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the presenter introduce herself at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the presenter introduce herself at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 18.0], "base_se_norm": [-0.9262503385543823, -0.7345011234283447], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.8525006175041199, -0.8230007290840149], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002492.pt"}
{"idx": 2494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TGrF7g4Z8EE.mp4", "duration": 135.59379166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person hold two different globes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person hold two different globes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 37.0], "base_se_norm": [-0.5427519083023071, -0.45425230264663696], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.5280020236968994, -0.45425230264663696], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002494.pt"}
{"idx": 2496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TGrF7g4Z8EE.mp4", "duration": 135.59379166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the presenter successfully levitates a ping pong ball with a hair dryer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the presenter successfully levitates a ping pong ball with a hair dryer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 115.0], "base_se_norm": [0.5339934825897217, 0.6962428092956543], "gt_se_raw": [106.0, 115.0], "gt_se_norm": [0.5634933710098267, 0.6962428092956543], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002496.pt"}
{"idx": 2498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfTfAIe9y84.mp4", "duration": 159.52603333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man in the glasses stumbles and falls to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man in the glasses stumbles and falls to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.8119429349899292, -0.7241829633712769], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.8119429349899292, -0.7492572069168091], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002498.pt"}
{"idx": 2500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfTfAIe9y84.mp4", "duration": 159.52603333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man and a woman are shown getting married with petals falling around them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man and a woman are shown getting married with petals falling around them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [-0.3731430768966675, -0.33553165197372437], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.3606058955192566, -0.33553165197372437], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002500.pt"}
{"idx": 2502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E_Ci-pAL4eE.mp4", "duration": 150.14999999999998, "prompt_text": "Please find the visual event described by the sentence 'When is the boy rolled up in a rug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy rolled up in a rug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.7735597491264343, -0.7335997223854065], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.8001998066902161, -0.7602397203445435], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002502.pt"}
{"idx": 2504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E_Ci-pAL4eE.mp4", "duration": 150.14999999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the mother and son baking a birthday cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the mother and son baking a birthday cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.2807192802429199, -0.22743922472000122], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.26739925146102905, -0.22743922472000122], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002504.pt"}
{"idx": 2506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E_Ci-pAL4eE.mp4", "duration": 150.14999999999998, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boy meets a dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boy meets a dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 110.0], "base_se_norm": [0.42524147033691406, 0.4652014970779419], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.42524147033691406, 0.4518815279006958], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002506.pt"}
{"idx": 2508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jXSX84eVUCs.mp4", "duration": 148.52, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a blue cricket appears on a rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a blue cricket appears on a rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.5556154251098633, -0.5017505884170532], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.5960140228271484, -0.5556154251098633], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002508.pt"}
{"idx": 2510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jXSX84eVUCs.mp4", "duration": 148.52, "prompt_text": "Please find the visual event described by the sentence 'When does the red dragon bite a man's butt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the red dragon bite a man's butt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 148.0], "base_se_norm": [0.9391326904296875, 0.9929975271224976], "gt_se_raw": [109.0, 111.0], "gt_se_norm": [0.4678157567977905, 0.4947481155395508], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002510.pt"}
{"idx": 2512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3EBfr6KdnQc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a dental professional using a tool in a patient's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a dental professional using a tool in a patient's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.672100305557251], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002512.pt"}
{"idx": 2514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3EBfr6KdnQc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a student is looking at papers instead of the patient.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a student is looking at papers instead of the patient.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.049278974533081055, 0.3771786689758301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002514.pt"}
{"idx": 2516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3EBfr6KdnQc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the motivational phrase 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the motivational phrase 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.508338451385498, 0.8362381458282471], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.508338451385498, 0.8362381458282471], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002516.pt"}
{"idx": 2518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3Qp1uk9zSo8.mp4", "duration": 136.59479166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the suit smile and give a thumbs-up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the suit smile and give a thumbs-up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 55.0], "base_se_norm": [-0.32647502422332764, -0.19469839334487915], "gt_se_raw": [46.0, 53.0], "gt_se_norm": [-0.32647502422332764, -0.2239820957183838], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002518.pt"}
{"idx": 2520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3Qp1uk9zSo8.mp4", "duration": 136.59479166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black suit perform a joyful, skipping dance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black suit perform a joyful, skipping dance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 84.0], "base_se_norm": [0.12742221355438232, 0.22991514205932617], "gt_se_raw": [77.0, 84.0], "gt_se_norm": [0.12742221355438232, 0.22991514205932617], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002520.pt"}
{"idx": 2522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DWynJkN5HbQ.mp4", "duration": 139.26412499999998, "prompt_text": "Please find the visual event described by the sentence 'When does the woman stretch her leg against the fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman stretch her leg against the fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.8707492351531982], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.8851103782653809], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002522.pt"}
{"idx": 2524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DWynJkN5HbQ.mp4", "duration": 139.26412499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man look shocked and bewildered?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man look shocked and bewildered?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.6840535998344421, -0.64096999168396], "gt_se_raw": [79.0, 85.0], "gt_se_norm": [0.1345348358154297, 0.2207019329071045], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002524.pt"}
{"idx": 2526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DWynJkN5HbQ.mp4", "duration": 139.26412499999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man, looking bewildered, walks away from the fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man, looking bewildered, walks away from the fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 115.0], "base_se_norm": [0.5940932035446167, 0.6515380144119263], "gt_se_raw": [111.0, 115.0], "gt_se_norm": [0.5940932035446167, 0.6515380144119263], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002526.pt"}
{"idx": 2528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vJG698U2Mvo.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the gorilla stops in the middle and beats its chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the gorilla stops in the middle and beats its chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.3386262059211731, -0.26514023542404175], "gt_se_raw": [27.0, 28.0], "gt_se_norm": [-0.3386262059211731, -0.31413090229034424], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002528.pt"}
{"idx": 2530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vJG698U2Mvo.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the black shirt and jeans on the left leave the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the black shirt and jeans on the left leave the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.6080747842788696, -0.46110284328460693], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [-0.1916542649269104, -0.1426635980606079], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002530.pt"}
{"idx": 2532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTLCo4PHRAw.mp4", "duration": 36.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a match being struck and ignited?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a match being struck and ignited?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 19.0], "base_se_norm": [-0.389454185962677, 0.05457913875579834], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.389454185962677, -0.27844589948654175], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002532.pt"}
{"idx": 2534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTLCo4PHRAw.mp4", "duration": 36.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the candle is lit and the match has been moved away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the candle is lit and the match has been moved away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 35.0], "base_se_norm": [0.33209991455078125, 0.9426456689834595], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [0.38760411739349365, 0.6096206903457642], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002534.pt"}
{"idx": 2536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTLCo4PHRAw.mp4", "duration": 36.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the period where the candle burns by itself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the period where the candle burns by itself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [0.38760411739349365, 0.9426456689834595], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [0.6651248931884766, 0.9426456689834595], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002536.pt"}
{"idx": 2538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aMtY7TqY_eQ.mp4", "duration": 12.5, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the pastry is being cut with a knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the pastry is being cut with a knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.36000001430511475, 0.12000000476837158], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.36000001430511475, 0.12000000476837158], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002538.pt"}
{"idx": 2540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aMtY7TqY_eQ.mp4", "duration": 12.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the text 'WE THOUGHT SO.' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the text 'WE THOUGHT SO.' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [0.12000000476837158, 0.6000000238418579], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [0.2799999713897705, 0.6000000238418579], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002540.pt"}
{"idx": 2542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/at-2rA44S-M.mp4", "duration": 52.452400000000004, "prompt_text": "Please find the visual event described by the sentence 'When can we see the exterior of the airplane hangar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the exterior of the airplane hangar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.35179322957992554, -0.1992740035057068], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.35179322957992554, -0.1611441969871521], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002542.pt"}
{"idx": 2544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/at-2rA44S-M.mp4", "duration": 52.452400000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man exclaim in pain after getting hurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man exclaim in pain after getting hurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [0.410802960395813, 0.5633221864700317], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [-0.00862497091293335, 0.029504895210266113], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002544.pt"}
{"idx": 2546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/at-2rA44S-M.mp4", "duration": 52.452400000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the animated character appear as an overlay on the black and white video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the animated character appear as an overlay on the black and white video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 52.0], "base_se_norm": [0.5633221864700317, 0.9827500581741333], "gt_se_raw": [41.0, 52.0], "gt_se_norm": [0.5633221864700317, 0.9827500581741333], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002546.pt"}
{"idx": 2548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/woWiyBgp5cs.mp4", "duration": 66.29956666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man start guiding the boy down the hallway by holding his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man start guiding the boy down the hallway by holding his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.45701003074645996, -0.3363456130027771], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.45701003074645996, -0.3665117025375366], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002548.pt"}
{"idx": 2550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/woWiyBgp5cs.mp4", "duration": 66.29956666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the pink egg toy is being held in a hand before it is hidden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the pink egg toy is being held in a hand before it is hidden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9698339104652405, -0.8190033435821533], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.9698339104652405, -0.8491694331169128], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002550.pt"}
{"idx": 2552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvhlK0WGBr4.mp4", "duration": 179.5, "prompt_text": "Please find the visual event described by the sentence 'When can we see a white van driving through the desert?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a white van driving through the desert?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.9220055937767029, -0.8885793685913086], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.9220055937767029, -0.8885793685913086], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002552.pt"}
{"idx": 2554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvhlK0WGBr4.mp4", "duration": 179.5, "prompt_text": "Please find the visual event described by the sentence 'When does the van stop next to the man lying on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the van stop next to the man lying on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 68.0], "base_se_norm": [-0.3426184058189392, -0.2423398494720459], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.3426184058189392, -0.2980501651763916], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002554.pt"}
{"idx": 2556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvhlK0WGBr4.mp4", "duration": 179.5, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man lying on a bench at what looks like a bus stop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man lying on a bench at what looks like a bus stop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 141.0], "base_se_norm": [0.43732595443725586, 0.5710306167602539], "gt_se_raw": [129.0, 138.0], "gt_se_norm": [0.43732595443725586, 0.5376044511795044], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002556.pt"}
{"idx": 2558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Qdwl-L_PR8.mp4", "duration": 160.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a person is rock climbing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a person is rock climbing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 111.0], "base_se_norm": [0.3297431468963623, 0.379453182220459], "gt_se_raw": [113.0, 116.0], "gt_se_norm": [0.4043081998825073, 0.44159066677093506], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002558.pt"}
{"idx": 2560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Qdwl-L_PR8.mp4", "duration": 160.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where two girls are washing dishes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where two girls are washing dishes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.5401822328567505, -0.5028997659683228], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.5153272747993469, -0.4904722571372986], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002560.pt"}
{"idx": 2562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hsm4poTWjMs.mp4", "duration": 252.25199999999998, "prompt_text": "Please find the visual event described by the sentence 'When are the two women lying on the sand surrounded by snakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women lying on the sand surrounded by snakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 55.0], "base_se_norm": [-0.6828567981719971, -0.5639281272888184], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.6749282479286194, -0.6590710878372192], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002562.pt"}
{"idx": 2564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hsm4poTWjMs.mp4", "duration": 252.25199999999998, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman in the leopard print outfit walk down the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman in the leopard print outfit walk down the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 118.0], "base_se_norm": [-0.159570574760437, -0.06442761421203613], "gt_se_raw": [106.0, 109.0], "gt_se_norm": [-0.159570574760437, -0.13578486442565918], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002564.pt"}
{"idx": 2566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hsm4poTWjMs.mp4", "duration": 252.25199999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the zebra-print outfit crawl on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 166.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the zebra-print outfit crawl on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 166.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 166.0], "base_se_norm": [0.27650129795074463, 0.3161442279815674], "gt_se_raw": [161.0, 165.0], "gt_se_norm": [0.27650129795074463, 0.3082156181335449], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002566.pt"}
{"idx": 2568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQP1duR2tf4.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'When does the man enter the room and start talking to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man enter the room and start talking to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 50.0], "base_se_norm": [-0.48645728826522827, -0.19758951663970947], "gt_se_raw": [32.0, 42.0], "gt_se_norm": [-0.48645728826522827, -0.325975239276886], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002568.pt"}
{"idx": 2570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQP1duR2tf4.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'When does the man admit that he was with the model and tries to downplay the incident?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man admit that he was with the model and tries to downplay the incident?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 50.0], "base_se_norm": [-0.30992698669433594, -0.19758951663970947], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [0.059181809425354004, 0.13942289352416992], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002570.pt"}
{"idx": 2572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oVS7kHYlLBc.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'When are the three sections of the pot being separated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three sections of the pot being separated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.9000499844551086, -0.7751124501228333], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.8750624656677246, -0.7751124501228333], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002572.pt"}
{"idx": 2574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oVS7kHYlLBc.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where water is being poured into the lower chamber of the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where water is being poured into the lower chamber of the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.6251873970031738, -0.4752624034881592], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [-0.6001999378204346, -0.4752624034881592], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002574.pt"}
{"idx": 2576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oVS7kHYlLBc.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the brewed coffee is served into a white cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the brewed coffee is served into a white cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [0.17441272735595703, 0.2993502616882324], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [0.17441272735595703, 0.2993502616882324], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002576.pt"}
{"idx": 2578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R-t2Zbzl5mk.mp4", "duration": 56.72332222222222, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when elbow macaroni is poured into the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when elbow macaroni is poured into the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.18904608488082886, -0.11852836608886719], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.18904608488082886, -0.11852836608886719], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002578.pt"}
{"idx": 2580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R-t2Zbzl5mk.mp4", "duration": 56.72332222222222, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the dish is being baked in the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the dish is being baked in the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [0.5161311626434326, 0.6924254894256592], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.5513900518417358, 0.657166600227356], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002580.pt"}
{"idx": 2582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oP0OgYPSXSE.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the photo of the young boy in the purple outfit displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the photo of the young boy in the purple outfit displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.6573548316955566, -0.5202967524528503], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.6573548316955566, -0.5431398153305054], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002582.pt"}
{"idx": 2584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oP0OgYPSXSE.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the picture where a boy in a yellow t-shirt is sitting next to a toucan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the picture where a boy in a yellow t-shirt is sitting next to a toucan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 47.0], "base_se_norm": [-0.06343650817871094, 0.07362151145935059], "gt_se_raw": [41.0, 47.0], "gt_se_norm": [-0.06343650817871094, 0.07362151145935059], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002584.pt"}
{"idx": 2586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oP0OgYPSXSE.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman, identified as Dee, is introduced with a caption.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman, identified as Dee, is introduced with a caption.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 15.0], "base_se_norm": [-0.9314709901809692, -0.6573548316955566], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.9314709901809692, -0.8172559142112732], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002586.pt"}
{"idx": 2588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkEEO_TLNbM.mp4", "duration": 65.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a model of the Space Shuttle Endeavour from the side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a model of the Space Shuttle Endeavour from the side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.08906883001327515, -0.02834010124206543], "gt_se_raw": [31.0, 32.0], "gt_se_norm": [-0.058704495429992676, -0.02834010124206543], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002588.pt"}
{"idx": 2590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkEEO_TLNbM.mp4", "duration": 65.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where children are reaching into a shallow water tank to touch rocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where children are reaching into a shallow water tank to touch rocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [-0.3319838047027588, -0.2712550759315491], "gt_se_raw": [23.0, 24.0], "gt_se_norm": [-0.3016194701194763, -0.2712550759315491], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002590.pt"}
{"idx": 2592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mize843q3iM.mp4", "duration": 30.113416666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are three female students observing a demonstration on a mannequin's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are three female students observing a demonstration on a mannequin's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.7343376874923706, -0.6015065312385559], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8007532358169556, -0.6679221391677856], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002592.pt"}
{"idx": 2594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mize843q3iM.mp4", "duration": 30.113416666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman raising her hands in celebration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman raising her hands in celebration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.2694286108016968, -0.13659745454788208], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.2694286108016968, -0.13659745454788208], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002594.pt"}
{"idx": 2596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mize843q3iM.mp4", "duration": 30.113416666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a student wearing blue gloves is drawing blood from a mannequin's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a student wearing blue gloves is drawing blood from a mannequin's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.7343376874923706, -0.6015065312385559], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.003766298294067383, 0.12906479835510254], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002596.pt"}
{"idx": 2598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hzkbhZhgHAU.mp4", "duration": 132.5324, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the orange octopus hangs onto the back of the moving truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the orange octopus hangs onto the back of the moving truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 53.0], "base_se_norm": [-0.36619269847869873, -0.2001955509185791], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.4114646315574646, -0.38128334283828735], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002598.pt"}
{"idx": 2600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hzkbhZhgHAU.mp4", "duration": 132.5324, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when both octopuses fall into the swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when both octopuses fall into the swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 85.0], "base_se_norm": [0.2223426103591919, 0.28270530700683594], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [0.2374333143234253, 0.26761460304260254], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002600.pt"}
{"idx": 2602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JvQcabZ1zrk.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man smoking a cigarette while wearing headphones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man smoking a cigarette while wearing headphones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [-0.4096190929412842, -0.2915428876876831], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.31778204441070557, -0.2915428876876831], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002602.pt"}
{"idx": 2604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JvQcabZ1zrk.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a red sweater post invitations on the event board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a red sweater post invitations on the event board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 108.0], "base_se_norm": [0.18076181411743164, 0.4169142246246338], "gt_se_raw": [90.0, 96.0], "gt_se_norm": [0.18076181411743164, 0.259479284286499], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002604.pt"}
{"idx": 2606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JvQcabZ1zrk.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the striped shirt angrily pull down the window blinds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the striped shirt angrily pull down the window blinds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [0.08892476558685303, 0.11516392230987549], "gt_se_raw": [84.0, 85.0], "gt_se_norm": [0.10204434394836426, 0.11516392230987549], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002606.pt"}
{"idx": 2608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p4cc4LkJFjg.mp4", "duration": 201.23436666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the speaker describes being perceived as a 'brown woman' and a 'foreign migrant'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the speaker describes being perceived as a 'brown woman' and a 'foreign migrant'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 52.0], "base_se_norm": [-0.5726375579833984, -0.4831896424293518], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7614721655845642, -0.7316561937332153], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002608.pt"}
{"idx": 2610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p4cc4LkJFjg.mp4", "duration": 201.23436666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker criticizes that pamphlets for immigration rights are in English.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker criticizes that pamphlets for immigration rights are in English.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 52.0], "base_se_norm": [-0.5726375579833984, -0.4831896424293518], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.3937417268753052, -0.35398703813552856], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002610.pt"}
{"idx": 2612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pEQ7QttNnxE.mp4", "duration": 25.4, "prompt_text": "Please find the visual event described by the sentence 'When is the person using a paintbrush-like tool near the light box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person using a paintbrush-like tool near the light box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.6062992215156555], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.8425197005271912], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002612.pt"}
{"idx": 2614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pEQ7QttNnxE.mp4", "duration": 25.4, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the person places a blue and a red object onto the light box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the person places a blue and a red object onto the light box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.6062992215156555, -0.3700787425041199], "gt_se_raw": [6.0, 10.0], "gt_se_norm": [-0.5275590419769287, -0.21259844303131104], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002614.pt"}
{"idx": 2616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pEQ7QttNnxE.mp4", "duration": 25.4, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the person is rubbing the surface with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the person is rubbing the surface with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [0.33858275413513184, 0.8110235929489136], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [0.33858275413513184, 0.8110235929489136], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002616.pt"}
{"idx": 2618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xZfZ-HB2yJI.mp4", "duration": 45.845800000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the child covers their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the child covers their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 43.0], "base_se_norm": [-0.3020080327987671, 0.875853419303894], "gt_se_raw": [16.0, 43.0], "gt_se_norm": [-0.3020080327987671, 0.875853419303894], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002618.pt"}
{"idx": 2620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xZfZ-HB2yJI.mp4", "duration": 45.845800000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the child in the purple shirt lets out a deep sigh.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the child in the purple shirt lets out a deep sigh.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.3020080327987671, -0.12751001119613647], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.6510040163993835, -0.5637550354003906], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002620.pt"}
{"idx": 2622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2kUMAA9yZgk.mp4", "duration": 131.88, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the girl running happily in the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the girl running happily in the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 13.0], "base_se_norm": [-0.9241734743118286, -0.8028510808944702], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8635122776031494, -0.8028510808944702], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002622.pt"}
{"idx": 2624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2kUMAA9yZgk.mp4", "duration": 131.88, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the girl hides behind a rock after seeing the little monkey.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the girl hides behind a rock after seeing the little monkey.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.21140432357788086, -0.15074312686920166], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [-0.24173492193222046, -0.19623905420303345], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002624.pt"}
{"idx": 2626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2kUMAA9yZgk.mp4", "duration": 131.88, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the excavator is cutting down trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the excavator is cutting down trees.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 103.0], "base_se_norm": [0.22838938236236572, 0.5620260238647461], "gt_se_raw": [96.0, 101.0], "gt_se_norm": [0.4558689594268799, 0.5316953659057617], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002626.pt"}
{"idx": 2628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/36OitAr_YQY.mp4", "duration": 47.98126666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man in the gray T-shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man in the gray T-shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.4164389371871948, -0.20802420377731323], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.37475597858428955, -0.20802420377731323], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002628.pt"}
{"idx": 2630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/36OitAr_YQY.mp4", "duration": 47.98126666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in black clothes talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in black clothes talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [0.04207336902618408, 0.5839515924453735], "gt_se_raw": [25.0, 30.0], "gt_se_norm": [0.04207336902618408, 0.2504880428314209], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002630.pt"}
{"idx": 2632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4wEA99kE8c.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the names of the three waterfalls that make up Niagara Falls presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the names of the three waterfalls that make up Niagara Falls presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 25.0], "base_se_norm": [-0.5895372629165649, -0.3963782787322998], "gt_se_raw": [17.0, 25.0], "gt_se_norm": [-0.5895372629165649, -0.3963782787322998], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002632.pt"}
{"idx": 2634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4wEA99kE8c.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the video visualizes the falls being frozen in winter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the video visualizes the falls being frozen in winter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 61.0], "base_se_norm": [0.25553321838378906, 0.472836971282959], "gt_se_raw": [54.0, 61.0], "gt_se_norm": [0.3038228750228882, 0.472836971282959], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002634.pt"}
{"idx": 2636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4nJYJ-QcABs.mp4", "duration": 203.36983333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the blonde girl in the green shirt shown up close for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the blonde girl in the green shirt shown up close for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.862319827079773, -0.8131482601165771], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.862319827079773, -0.8229826092720032], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002636.pt"}
{"idx": 2638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4nJYJ-QcABs.mp4", "duration": 203.36983333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the instruction 'SIT IN SILENCE AND STILLNESS' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the instruction 'SIT IN SILENCE AND STILLNESS' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.8819884061813354, -0.8524854779243469], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.8918226957321167, -0.862319827079773], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002638.pt"}
{"idx": 2640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4nJYJ-QcABs.mp4", "duration": 203.36983333333333, "prompt_text": "Please find the visual event described by the sentence 'When do the children open their eyes after the meditation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 194.0 - 202.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the children open their eyes after the meditation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 194.0 - 202.0 seconds.<|im_end|>\n", "base_se_raw": [194.0, 202.0], "base_se_norm": [0.9078543186187744, 0.986528754234314], "gt_se_raw": [193.0, 199.0], "gt_se_norm": [0.8980200290679932, 0.9570257663726807], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002640.pt"}
{"idx": 2642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FatA6Pyw_D8.mp4", "duration": 72.0, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the dog shakes its body rapidly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the dog shakes its body rapidly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.25, -0.1388888955116272], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [-0.3888888955116272, -0.3611111044883728], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002642.pt"}
{"idx": 2644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FatA6Pyw_D8.mp4", "duration": 72.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the cat holds up a stick for the dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the cat holds up a stick for the dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [0.2222222089767456, 0.4166666269302368], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [0.2222222089767456, 0.4166666269302368], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002644.pt"}
{"idx": 2646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZHlfQSL7dJ8.mp4", "duration": 57.22383333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a group of people taking a photo together, including a baby in a stroller?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a group of people taking a photo together, including a baby in a stroller?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.7553467154502869, -0.5456438660621643], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.7553467154502869, -0.5456438660621643], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002646.pt"}
{"idx": 2648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZHlfQSL7dJ8.mp4", "duration": 57.22383333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is a man in a blue suit and red shirt smiling while standing at a podium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a man in a blue suit and red shirt smiling while standing at a podium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.021386802196502686, 0.15336549282073975], "gt_se_raw": [27.0, 33.0], "gt_se_norm": [-0.05633729696273804, 0.15336549282073975], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002648.pt"}
{"idx": 2650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZHlfQSL7dJ8.mp4", "duration": 57.22383333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where two men are smiling together for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where two men are smiling together for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [0.46791982650756836, 0.5378206968307495], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [0.46791982650756836, 0.5727711915969849], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002650.pt"}
{"idx": 2652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-TQpVg4vcrk.mp4", "duration": 60.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the two men shake hands for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the two men shake hands for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.7336292862892151, -0.6670366525650024], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.7336292862892151, -0.6670366525650024], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002652.pt"}
{"idx": 2654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-TQpVg4vcrk.mp4", "duration": 60.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman in the red sweater and black beanie laughs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman in the red sweater and black beanie laughs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 47.0], "base_se_norm": [0.49833524227142334, 0.5649279356002808], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.531631588935852, 0.5982241630554199], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002654.pt"}
{"idx": 2656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f25JmqQMYow.mp4", "duration": 115.6155, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the newborn gorilla resting in its mother's embrace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the newborn gorilla resting in its mother's embrace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, -0.7232204675674438], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9308051466941833], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002656.pt"}
{"idx": 2658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f25JmqQMYow.mp4", "duration": 115.6155, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the mother gorilla walks while carrying her baby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the mother gorilla walks while carrying her baby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.49833714962005615, -0.4291422963142395], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.49833714962005615, -0.4291422963142395], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002658.pt"}
{"idx": 2660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f25JmqQMYow.mp4", "duration": 115.6155, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a gorilla starts tearing open a polka-dotted gift box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a gorilla starts tearing open a polka-dotted gift box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.22155767679214478, -0.10046666860580444], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [-0.22155767679214478, -0.15236282348632812], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002660.pt"}
{"idx": 2662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5hpBAn5lQPs.mp4", "duration": 57.057, "prompt_text": "Please find the visual event described by the sentence 'When is a boy in a wheelchair shown celebrating with a 'Cougar Pride' award?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a boy in a wheelchair shown celebrating with a 'Cougar Pride' award?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 43.0], "base_se_norm": [-0.7546313405036926, 0.5072647333145142], "gt_se_raw": [7.0, 43.0], "gt_se_norm": [-0.7546313405036926, 0.5072647333145142], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002662.pt"}
{"idx": 2664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5hpBAn5lQPs.mp4", "duration": 57.057, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where two boys are relaxing on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where two boys are relaxing on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 56.0], "base_se_norm": [0.5423173904418945, 0.9629493951797485], "gt_se_raw": [43.0, 57.0], "gt_se_norm": [0.5072647333145142, 0.9980020523071289], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002664.pt"}
{"idx": 2666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fTcSVQJ2h8g.mp4", "duration": 184.35083333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where two different final prices are presented for the car purchase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where two different final prices are presented for the car purchase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 52.0], "base_se_norm": [-0.57689368724823, -0.43585824966430664], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.57689368724823, -0.5334981679916382], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002666.pt"}
{"idx": 2668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fTcSVQJ2h8g.mp4", "duration": 184.35083333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the text 'BLACK PEOPLE ARE INCARCERATED AT NEARLY SIX TIMES THE RATE OF WHITE PEOPLE' is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the text 'BLACK PEOPLE ARE INCARCERATED AT NEARLY SIX TIMES THE RATE OF WHITE PEOPLE' is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 114.0], "base_se_norm": [0.16083014011383057, 0.2367722988128662], "gt_se_raw": [106.0, 114.0], "gt_se_norm": [0.14998126029968262, 0.2367722988128662], "start_token_span": [83, 88], "end_token_span": [90, 95], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002668.pt"}
{"idx": 2670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fTcSVQJ2h8g.mp4", "duration": 184.35083333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the scene where real estate agents are showing houses to the two men.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the scene where real estate agents are showing houses to the two men.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 91.0], "base_se_norm": [-0.13208961486816406, -0.012751936912536621], "gt_se_raw": [80.0, 86.0], "gt_se_norm": [-0.13208961486816406, -0.06699633598327637], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002670.pt"}
{"idx": 2672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_QG77WomHo.mp4", "duration": 159.159, "prompt_text": "Please find the visual event described by the sentence 'When are workers assembling scaffolding on the side of the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are workers assembling scaffolding on the side of the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.6984148025512695, -0.5978863835334778], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.6984148025512695, -0.6355845332145691], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002672.pt"}
{"idx": 2674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_QG77WomHo.mp4", "duration": 159.159, "prompt_text": "Please find the visual event described by the sentence 'At what point do people raise their glasses for a toast on the rooftop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do people raise their glasses for a toast on the rooftop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 75.0], "base_se_norm": [-0.08267831802368164, -0.05754619836807251], "gt_se_raw": [73.0, 75.0], "gt_se_norm": [-0.08267831802368164, -0.05754619836807251], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002674.pt"}
{"idx": 2676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ywzm6kvDo7o.mp4", "duration": 135.76896666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the presenter standing with both palms facing up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the presenter standing with both palms facing up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.8379600048065186, -0.7937673330307007], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [-0.5138800740242004, -0.48441827297210693], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002676.pt"}
{"idx": 2678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ywzm6kvDo7o.mp4", "duration": 135.76896666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the spiny lobster appear from under the rocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the spiny lobster appear from under the rocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 67.0], "base_se_norm": [-0.057221949100494385, -0.013029217720031738], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [-0.057221949100494385, -0.013029217720031738], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002678.pt"}
{"idx": 2680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ywzm6kvDo7o.mp4", "duration": 135.76896666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the leopard shark is swimming across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the leopard shark is swimming across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 134.0], "base_se_norm": [0.8413633108139038, 0.9739415645599365], "gt_se_raw": [125.0, 134.0], "gt_se_norm": [0.8413633108139038, 0.9739415645599365], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002680.pt"}
{"idx": 2682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SHR0GuIf5eM.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a woman in a blue polo shirt is talking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a woman in a blue polo shirt is talking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.4753605127334595, -0.27862071990966797], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002682.pt"}
{"idx": 2684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SHR0GuIf5eM.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman with safety glasses working on the internal parts of an engine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman with safety glasses working on the internal parts of an engine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002684.pt"}
{"idx": 2686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XlUMyKgmy6k.mp4", "duration": 133.24, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when two men are getting married.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when two men are getting married.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 120.0], "base_se_norm": [0.7562292814254761, 0.8012608289718628], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.7562292814254761, 0.8012608289718628], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002686.pt"}
{"idx": 2688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XlUMyKgmy6k.mp4", "duration": 133.24, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a magical pony is flying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a magical pony is flying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 108.0], "base_se_norm": [0.531071662902832, 0.6211347579956055], "gt_se_raw": [103.0, 105.0], "gt_se_norm": [0.5460821390151978, 0.5761032104492188], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002688.pt"}
{"idx": 2690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XlUMyKgmy6k.mp4", "duration": 133.24, "prompt_text": "Please find the visual event described by the sentence 'Find the time range when a dinosaur appears on the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range when a dinosaur appears on the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [-0.32452720403671265, -0.2794956564903259], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [-0.30951666831970215, -0.2794956564903259], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002690.pt"}
{"idx": 2692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/19143I5f3Lg.mp4", "duration": 150.88406666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the definition of PBIS is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the definition of PBIS is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 58.0], "base_se_norm": [-0.33723950386047363, -0.2311977744102478], "gt_se_raw": [49.0, 58.0], "gt_se_norm": [-0.3504946827888489, -0.2311977744102478], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002692.pt"}
{"idx": 2694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/19143I5f3Lg.mp4", "duration": 150.88406666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where several teachers and Kelsey are sitting around a table in a meeting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where several teachers and Kelsey are sitting around a table in a meeting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 110.0], "base_se_norm": [0.3652867078781128, 0.45807313919067383], "gt_se_raw": [103.0, 110.0], "gt_se_norm": [0.3652867078781128, 0.45807313919067383], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002694.pt"}
{"idx": 2696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Hf_2ilitep0.mp4", "duration": 117.8177, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the background laughing heartily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the background laughing heartily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.6265417337417603, -0.5246894359588623], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.6265417337417603, -0.5416648387908936], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002696.pt"}
{"idx": 2698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Hf_2ilitep0.mp4", "duration": 117.8177, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child enthusiastically strum the guitar for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child enthusiastically strum the guitar for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.5925909280776978, -0.5246894359588623], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [-0.6435170769691467, -0.5416648387908936], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002698.pt"}
{"idx": 2700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Hf_2ilitep0.mp4", "duration": 117.8177, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the conversation about attempting to play a chord?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the conversation about attempting to play a chord?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 68.0], "base_se_norm": [-0.2700587511062622, 0.1543257236480713], "gt_se_raw": [34.0, 43.0], "gt_se_norm": [-0.42283713817596436, -0.2700587511062622], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002700.pt"}
{"idx": 2702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1O-HcBYpNM0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman guides the girl's hands to interact with the spiky ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman guides the girl's hands to interact with the spiky ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.6126651763916016], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.6126651763916016], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002702.pt"}
{"idx": 2704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1O-HcBYpNM0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the girl picks up the wooden block and places it in the woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the girl picks up the wooden block and places it in the woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [0.25883805751800537, 0.4525054693222046], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.25883805751800537, 0.4525054693222046], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002704.pt"}
{"idx": 2706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kQQaEo8MqDQ.mp4", "duration": 261.96666666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the white shirt touch the painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the white shirt touch the painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 24.0], "base_se_norm": [-0.8473088145256042, -0.8167705535888672], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.8549433946609497, -0.8167705535888672], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002706.pt"}
{"idx": 2708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kQQaEo8MqDQ.mp4", "duration": 261.96666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman express that she can 'see' the Mona Lisa without her eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman express that she can 'see' the Mona Lisa without her eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 77.0], "base_se_norm": [-0.5113882422447205, -0.4121389389038086], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [-0.618272066116333, -0.5953683853149414], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002708.pt"}
{"idx": 2710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kQQaEo8MqDQ.mp4", "duration": 261.96666666666664, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'Basic Civil Rights' shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 224.0 - 238.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'Basic Civil Rights' shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 224.0 - 238.0 seconds.<|im_end|>\n", "base_se_raw": [224.0, 238.0], "base_se_norm": [0.7101411819458008, 0.8170250654220581], "gt_se_raw": [223.0, 237.0], "gt_se_norm": [0.7025066614151001, 0.8093904256820679], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002710.pt"}
{"idx": 2712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CKwUNBEFI0E.mp4", "duration": 156.768, "prompt_text": "Please find the visual event described by the sentence 'A person kneels down and prepares the jet on the side of the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person kneels down and prepares the jet on the side of the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 66.0], "base_se_norm": [-0.4896917939186096, -0.1579914689064026], "gt_se_raw": [42.0, 67.0], "gt_se_norm": [-0.46417635679244995, -0.14523375034332275], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002712.pt"}
{"idx": 2714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CKwUNBEFI0E.mp4", "duration": 156.768, "prompt_text": "Please find the visual event described by the sentence 'The jet is performing a barrel roll in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The jet is performing a barrel roll in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 114.0], "base_se_norm": [0.3523167371749878, 0.4543783664703369], "gt_se_raw": [96.0, 100.0], "gt_se_norm": [0.224739670753479, 0.27577054500579834], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002714.pt"}
{"idx": 2716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4XwDbm_37cw.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'When is an application used to identify the value of dollar bills?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is an application used to identify the value of dollar bills?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 83.0], "base_se_norm": [-0.07784527540206909, 0.020517945289611816], "gt_se_raw": [76.0, 83.0], "gt_se_norm": [-0.06554985046386719, 0.020517945289611816], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002716.pt"}
{"idx": 2718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4XwDbm_37cw.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'When does a voiceover announce the names of apps as a finger touches them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a voiceover announce the names of apps as a finger touches them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 71.0], "base_se_norm": [-0.20079922676086426, -0.12702685594558716], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.18850386142730713, -0.12702685594558716], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002718.pt"}
{"idx": 2720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4XwDbm_37cw.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'At what time does the video demonstrate how the Vision Simulator app works?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the video demonstrate how the Vision Simulator app works?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 83.0], "base_se_norm": [-0.07784527540206909, 0.020517945289611816], "gt_se_raw": [134.0, 148.0], "gt_se_norm": [0.6475831270217896, 0.8197187185287476], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002720.pt"}
{"idx": 2722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DMCSP73Rq4I.mp4", "duration": 57.182125, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man is sitting and reading the paper before he crumples it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man is sitting and reading the paper before he crumples it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, -0.23052877187728882], "gt_se_raw": [0.0, 22.0], "gt_se_norm": [-1.0, -0.23052877187728882], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002722.pt"}
{"idx": 2724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DMCSP73Rq4I.mp4", "duration": 57.182125, "prompt_text": "Please find the visual event described by the sentence 'Show the man holding the crumpled flyer and looking up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man holding the crumpled flyer and looking up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 30.0], "base_se_norm": [-0.23052877187728882, 0.049278974533081055], "gt_se_raw": [24.0, 31.0], "gt_se_norm": [-0.16057682037353516, 0.08425486087799072], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002724.pt"}
{"idx": 2726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGJ_td3ZUIU.mp4", "duration": 196.62976666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the speaker's name, Arielle Newton, displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the speaker's name, Arielle Newton, displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9898285865783691, -0.9186288118362427], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9898285865783691, -0.9186288118362427], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002726.pt"}
{"idx": 2728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGJ_td3ZUIU.mp4", "duration": 196.62976666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video pose the question, 'What about people who think talking about our differences divides us?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video pose the question, 'What about people who think talking about our differences divides us?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [-0.6440010070800781, -0.6033153533935547], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.6338295936584473, -0.6033153533935547], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002728.pt"}
{"idx": 2730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGJ_td3ZUIU.mp4", "duration": 196.62976666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the speaker is talking about unlearning the compartmentalizing of her being.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the speaker is talking about unlearning the compartmentalizing of her being.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 76.0], "base_se_norm": [-0.49142998456954956, -0.22697359323501587], "gt_se_raw": [146.0, 167.0], "gt_se_norm": [0.48502445220947266, 0.6986238956451416], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002730.pt"}
{"idx": 2732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hIC7XFLhJ7A.mp4", "duration": 106.37293333333334, "prompt_text": "Please find the visual event described by the sentence 'When is a woman wearing a black floral jacket speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman wearing a black floral jacket speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.8871893286705017, -0.7743786573410034], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.8871893286705017, -0.7743786573410034], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002732.pt"}
{"idx": 2734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hIC7XFLhJ7A.mp4", "duration": 106.37293333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a vintage dress hold a red folder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a vintage dress hold a red folder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 93.0], "base_se_norm": [0.579349160194397, 0.7485650777816772], "gt_se_raw": [83.0, 93.0], "gt_se_norm": [0.5605473518371582, 0.7485650777816772], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002734.pt"}
{"idx": 2736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/545Tp9xXEIQ.mp4", "duration": 46.08770833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the yellow minion laugh at the purple minion, who is on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the yellow minion laugh at the purple minion, who is on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 35.0], "base_se_norm": [0.17167901992797852, 0.5188431739807129], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.17167901992797852, 0.2584700584411621], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002736.pt"}
{"idx": 2738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/545Tp9xXEIQ.mp4", "duration": 46.08770833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do both minions get into a fighting stance before the chase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do both minions get into a fighting stance before the chase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 35.0], "base_se_norm": [0.12828350067138672, 0.5188431739807129], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.4320521354675293, 0.5188431739807129], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002738.pt"}
{"idx": 2740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/545Tp9xXEIQ.mp4", "duration": 46.08770833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the yellow minion wave hello?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the yellow minion wave hello?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 2.0], "base_se_norm": [-0.9566044807434082, -0.9132089614868164], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9566044807434082, -0.9132089614868164], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002740.pt"}
{"idx": 2742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fiKThScYotw.mp4", "duration": 60.3, "prompt_text": "Please find the visual event described by the sentence 'At what time does the phrase '...AND SIDEWALKS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the phrase '...AND SIDEWALKS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [-0.10447758436203003, -0.038142621517181396], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.10447758436203003, -0.038142621517181396], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002742.pt"}
{"idx": 2744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fiKThScYotw.mp4", "duration": 60.3, "prompt_text": "Please find the visual event described by the sentence 'When is the boy walking indoors on a tiled floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy walking indoors on a tiled floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 51.0], "base_se_norm": [0.16086232662200928, 0.6915422677993774], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.16086232662200928, 0.22719740867614746], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002744.pt"}
{"idx": 2746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_yQABXDAy2Q.mp4", "duration": 32.098733333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the video display its title, 'Learning Braille and Tools for Vision'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display its title, 'Learning Braille and Tools for Vision'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.688461184501648], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.7507689595222473], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002746.pt"}
{"idx": 2748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_yQABXDAy2Q.mp4", "duration": 32.098733333333335, "prompt_text": "Please find the visual event described by the sentence 'When are the benefits of Braille for reading and math mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the benefits of Braille for reading and math mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.433078408241272, 0.744617223739624], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.433078408241272, 0.744617223739624], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002748.pt"}
{"idx": 2750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_yQABXDAy2Q.mp4", "duration": 32.098733333333335, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the girl responds that learning Braille is hard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the girl responds that learning Braille is hard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.8069249391555786, 0.9315404891967773], "gt_se_raw": [30.0, 31.0], "gt_se_norm": [0.8692327737808228, 0.9315404891967773], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002750.pt"}
{"idx": 2752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NWIVHL9bPPU.mp4", "duration": 146.27944611277945, "prompt_text": "Please find the visual event described by the sentence 'When can we see cartoon illustrations of children with disabilities in various settings like school and with their families?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see cartoon illustrations of children with disabilities in various settings like school and with their families?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.671860933303833, -0.589826226234436], "gt_se_raw": [24.0, 30.0], "gt_se_norm": [-0.671860933303833, -0.589826226234436], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002752.pt"}
{"idx": 2754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NWIVHL9bPPU.mp4", "duration": 146.27944611277945, "prompt_text": "Please find the visual event described by the sentence 'Please provide the time range for when the signpost with words like 'HELP', 'SUPPORT', and 'ASSISTANCE' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please provide the time range for when the signpost with words like 'HELP', 'SUPPORT', and 'ASSISTANCE' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 91.0], "base_se_norm": [0.13481426239013672, 0.2441939115524292], "gt_se_raw": [83.0, 91.0], "gt_se_norm": [0.13481426239013672, 0.2441939115524292], "start_token_span": [83, 87], "end_token_span": [89, 93], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002754.pt"}
{"idx": 2756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p1a5oijzFaY.mp4", "duration": 111.54476666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the blue jacket place the milk cartons on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the blue jacket place the milk cartons on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.677259624004364, -0.5876095294952393], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.6593296527862549, -0.6055395603179932], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002756.pt"}
{"idx": 2758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p1a5oijzFaY.mp4", "duration": 111.54476666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman takes off her blue jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman takes off her blue jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [-0.26486915349960327, -0.19314908981323242], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [-0.24693912267684937, -0.19314908981323242], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002758.pt"}
{"idx": 2760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p1a5oijzFaY.mp4", "duration": 111.54476666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when the woman touches the other person's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when the woman touches the other person's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 38.0], "base_se_norm": [-0.5696794986724854, -0.318659245967865], "gt_se_raw": [91.0, 93.0], "gt_se_norm": [0.6316318511962891, 0.6674919128417969], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002760.pt"}
{"idx": 2762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODDZPV-Avfw.mp4", "duration": 345.64, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where two characters are embracing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 253.0 - 256.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where two characters are embracing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 253.0 - 256.0 seconds.<|im_end|>\n", "base_se_raw": [253.0, 256.0], "base_se_norm": [0.46395087242126465, 0.48131000995635986], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.855340838432312, -0.8437681198120117], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002762.pt"}
{"idx": 2764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODDZPV-Avfw.mp4", "duration": 345.64, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a character is lying on a pile of books and reading.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a character is lying on a pile of books and reading.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 135.0], "base_se_norm": [-0.2767041325569153, -0.21884042024612427], "gt_se_raw": [125.0, 127.0], "gt_se_norm": [-0.2767041325569153, -0.2651313543319702], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002764.pt"}
{"idx": 2766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vI_B7dtF7Q0.mp4", "duration": 297.04675, "prompt_text": "Please find the visual event described by the sentence 'When does the 'Ask Orange #63' title card appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'Ask Orange #63' title card appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 61.0], "base_se_norm": [-0.6094890832901001, -0.5892902612686157], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.6094890832901001, -0.5892902612686157], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002766.pt"}
{"idx": 2768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vI_B7dtF7Q0.mp4", "duration": 297.04675, "prompt_text": "Please find the visual event described by the sentence 'When is the donut chopped by an axe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the donut chopped by an axe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 138.0], "base_se_norm": [-0.08431923389434814, -0.0708533525466919], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [-0.08431923389434814, -0.0708533525466919], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002768.pt"}
{"idx": 2770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vI_B7dtF7Q0.mp4", "duration": 297.04675, "prompt_text": "Please find the visual event described by the sentence 'At what point does the UNO card light the TNT with a lighter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 272.0 - 278.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the UNO card light the TNT with a lighter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 272.0 - 278.0 seconds.<|im_end|>\n", "base_se_raw": [272.0, 278.0], "base_se_norm": [0.8313615322113037, 0.8717591762542725], "gt_se_raw": [276.0, 278.0], "gt_se_norm": [0.8582932949066162, 0.8717591762542725], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002770.pt"}
{"idx": 2772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6fan6ggvh4U.mp4", "duration": 178.2781, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the teacher is holding an apple and talking to a student in a red hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the teacher is holding an apple and talking to a student in a red hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 53.0], "base_se_norm": [-0.5400444865226746, -0.4054233431816101], "gt_se_raw": [43.0, 52.0], "gt_se_norm": [-0.5176076292991638, -0.4166417717933655], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002772.pt"}
{"idx": 2774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6fan6ggvh4U.mp4", "duration": 178.2781, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the character in brown pants is singing and dancing on the teacher's desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the character in brown pants is singing and dancing on the teacher's desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 157.0], "base_se_norm": [0.2676823139190674, 0.7612930536270142], "gt_se_raw": [113.0, 123.0], "gt_se_norm": [0.2676823139190674, 0.3798666000366211], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002774.pt"}
{"idx": 2776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUVwR0rw5fk.mp4", "duration": 6.881874999999999, "prompt_text": "Please find the visual event described by the sentence 'When is Homer raising a glass of beer while standing on a stack of barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Homer raising a glass of beer while standing on a stack of barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, 0.7437108755111694], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, 0.7437108755111694], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002776.pt"}
{"idx": 2778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUVwR0rw5fk.mp4", "duration": 6.881874999999999, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the speaker mentions 'the cause of and solution to all of life's problems'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the speaker mentions 'the cause of and solution to all of life's problems'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.12814456224441528, 0.4530923366546631], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.7093815207481384, 0.7437108755111694], "start_token_span": [77, 80], "end_token_span": [82, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002778.pt"}
{"idx": 2780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUVwR0rw5fk.mp4", "duration": 6.881874999999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the scene where a man with blue glasses and an orange shirt appears on the left side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the scene where a man with blue glasses and an orange shirt appears on the left side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, 0.7437108755111694], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, 0.7437108755111694], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002780.pt"}
{"idx": 2782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/63d0vV0kk_Y.mp4", "duration": 111.74496666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker explain how to find the UDL guidelines on the CAST website?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker explain how to find the UDL guidelines on the CAST website?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 53.0], "base_se_norm": [-0.2840840816497803, -0.05141139030456543], "gt_se_raw": [40.0, 51.0], "gt_se_norm": [-0.2840840816497803, -0.08720719814300537], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002782.pt"}
{"idx": 2784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/63d0vV0kk_Y.mp4", "duration": 111.74496666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a classroom bulletin board titled 'WHAT ARE WE LEARNING?' is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a classroom bulletin board titled 'WHAT ARE WE LEARNING?' is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 69.0], "base_se_norm": [0.12756752967834473, 0.23495495319366455], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [0.12756752967834473, 0.21705710887908936], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002784.pt"}
{"idx": 2786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WP_L1I4lmFE.mp4", "duration": 93.32, "prompt_text": "Please find the visual event described by the sentence 'When is the blue fish sleeping inside a brain-shaped coral?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the blue fish sleeping inside a brain-shaped coral?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.6785255074501038, -0.5713673233985901], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.6785255074501038, -0.5713673233985901], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002786.pt"}
{"idx": 2788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WP_L1I4lmFE.mp4", "duration": 93.32, "prompt_text": "Please find the visual event described by the sentence 'When does the blue fish say 'I remembered something'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue fish say 'I remembered something'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.14273464679718018, -0.07843977212905884], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [0.1787397861480713, 0.22160303592681885], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002788.pt"}
{"idx": 2790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WP_L1I4lmFE.mp4", "duration": 93.32, "prompt_text": "Please find the visual event described by the sentence 'When does the 'Finding Dory' title appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'Finding Dory' title appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [0.607372522354126, 0.6716673374176025], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.607372522354126, 0.6716673374176025], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002790.pt"}
{"idx": 2792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ipnGPeRIy2k.mp4", "duration": 228.7285, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where one of the old ladies is laughing and covering her face with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where one of the old ladies is laughing and covering her face with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 89.0], "base_se_norm": [-0.25676071643829346, -0.221784770488739], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [-0.25676071643829346, -0.23052877187728882], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002792.pt"}
{"idx": 2794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ipnGPeRIy2k.mp4", "duration": 228.7285, "prompt_text": "Please find the visual event described by the sentence 'When are the two ladies talking about Justin Bieber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two ladies talking about Justin Bieber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 153.0], "base_se_norm": [0.12797486782073975, 0.33783066272735596], "gt_se_raw": [129.0, 153.0], "gt_se_norm": [0.12797486782073975, 0.33783066272735596], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002794.pt"}
{"idx": 2796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bvylqPgr-bQ.mp4", "duration": 180.14663333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the briefcase shown next to the 'SECTION 15' text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the briefcase shown next to the 'SECTION 15' text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 50.0], "base_se_norm": [-0.48930490016937256, -0.4448966383934021], "gt_se_raw": [46.0, 51.0], "gt_se_norm": [-0.48930490016937256, -0.43379461765289307], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002796.pt"}
{"idx": 2798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bvylqPgr-bQ.mp4", "duration": 180.14663333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where James is told he's hired.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where James is told he's hired.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 95.0], "base_se_norm": [0.010288119316101074, 0.05469632148742676], "gt_se_raw": [91.0, 94.0], "gt_se_norm": [0.010288119316101074, 0.04359424114227295], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002798.pt"}
{"idx": 2800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bvylqPgr-bQ.mp4", "duration": 180.14663333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the 'HELP WANTED' sign appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'HELP WANTED' sign appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 73.0], "base_se_norm": [-0.2228553295135498, -0.18954914808273315], "gt_se_raw": [70.0, 73.0], "gt_se_norm": [-0.2228553295135498, -0.18954914808273315], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002800.pt"}
{"idx": 2802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pX7dCnUMpbk.mp4", "duration": 110.235125, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of a young man smiling with his eyes closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of a young man smiling with his eyes closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 16.0], "base_se_norm": [-0.8911417722702026, -0.7097114324569702], "gt_se_raw": [7.0, 16.0], "gt_se_norm": [-0.8729987144470215, -0.7097114324569702], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002802.pt"}
{"idx": 2804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pX7dCnUMpbk.mp4", "duration": 110.235125, "prompt_text": "Please find the visual event described by the sentence 'At what point is the young man walking indoors with a white cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the young man walking indoors with a white cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 50.0], "base_se_norm": [-0.32870757579803467, -0.09284812211990356], "gt_se_raw": [37.0, 50.0], "gt_se_norm": [-0.32870757579803467, -0.09284812211990356], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002804.pt"}
{"idx": 2806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZrO_qsIJw6o.mp4", "duration": 182.91624958291627, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man showing off his pectoral muscles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man showing off his pectoral muscles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 42.0], "base_se_norm": [-0.7047829031944275, -0.5407733917236328], "gt_se_raw": [28.0, 42.0], "gt_se_norm": [-0.6938489675521851, -0.5407733917236328], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002806.pt"}
{"idx": 2808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZrO_qsIJw6o.mp4", "duration": 182.91624958291627, "prompt_text": "Please find the visual event described by the sentence 'Show the footage of the extinct Tasmanian tiger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the footage of the extinct Tasmanian tiger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 98.0], "base_se_norm": [-0.15808463096618652, 0.07152867317199707], "gt_se_raw": [78.0, 98.0], "gt_se_norm": [-0.14715063571929932, 0.07152867317199707], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002808.pt"}
{"idx": 2810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZrO_qsIJw6o.mp4", "duration": 182.91624958291627, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the afterburner of a fighter jet is ignited.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the afterburner of a fighter jet is ignited.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 167.0], "base_se_norm": [0.6510288715362549, 0.825972318649292], "gt_se_raw": [150.0, 167.0], "gt_se_norm": [0.6400948762893677, 0.825972318649292], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002810.pt"}
{"idx": 2812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xLE1rO60uCk.mp4", "duration": 69.83643333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the man mentions his 'Bombardier Challenger'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the man mentions his 'Bombardier Challenger'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.5131480693817139, -0.4272330403327942], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.3413180112838745, -0.22676461935043335], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002812.pt"}
{"idx": 2814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xLE1rO60uCk.mp4", "duration": 69.83643333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the man talks about spending more time with his friends and family.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the man talks about spending more time with his friends and family.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 48.0], "base_se_norm": [0.0023421049118041992, 0.37464070320129395], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.05961883068084717, 0.11689555644989014], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002814.pt"}
{"idx": 2816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-9WnLz26Bwc.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the white moose drinking from the stream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the white moose drinking from the stream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.8530280590057373], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.902018666267395], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002816.pt"}
{"idx": 2818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-9WnLz26Bwc.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the moose get out of the water and shake its body dry?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the moose get out of the water and shake its body dry?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 17.0], "base_se_norm": [-0.902018666267395, -0.5835794806480408], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.6325701475143433, -0.5835794806480408], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002818.pt"}
{"idx": 2820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-9WnLz26Bwc.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the moose turn its head to look directly towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the moose turn its head to look directly towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.41211217641830444, -0.31413090229034424], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.3876168727874756, -0.2896355390548706], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002820.pt"}
{"idx": 2822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h56Cr2ho1Y8.mp4", "duration": 186.3528333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment showing a bicycle with a bag in its basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment showing a bicycle with a bag in its basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 168.0], "base_se_norm": [0.7922990322113037, 0.8030314445495605], "gt_se_raw": [166.0, 168.0], "gt_se_norm": [0.7815667390823364, 0.8030314445495605], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002822.pt"}
{"idx": 2824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h56Cr2ho1Y8.mp4", "duration": 186.3528333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a man and a woman are doing mountain climbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a man and a woman are doing mountain climbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 87.0], "base_se_norm": [-0.27020156383514404, -0.0662873387336731], "gt_se_raw": [69.0, 87.0], "gt_se_norm": [-0.25946927070617676, -0.0662873387336731], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002824.pt"}
{"idx": 2826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K-RVlLR7wMw.mp4", "duration": 29.9299, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the teacher in the striped shirt gesturing with both hands simultaneously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the teacher in the striped shirt gesturing with both hands simultaneously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.13130342960357666, 0.1359877586364746], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.13130342960357666, 0.00234222412109375], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002826.pt"}
{"idx": 2828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K-RVlLR7wMw.mp4", "duration": 29.9299, "prompt_text": "Please find the visual event described by the sentence 'When does the little girl briefly look down at the tablet in front of her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little girl briefly look down at the tablet in front of her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.6705702543258667, 0.8710386753082275], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.8042159080505371, 0.9378615617752075], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002828.pt"}
{"idx": 2830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K-RVlLR7wMw.mp4", "duration": 29.9299, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the seated teacher is signing with her right hand near her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the seated teacher is signing with her right hand near her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.732708752155304, -0.13130342960357666], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [0.6705702543258667, 0.8042159080505371], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002830.pt"}
{"idx": 2832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TvMAUk3Etjw.mp4", "duration": 187.4, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a player gets two trap kills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a player gets two trap kills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [-0.7225186824798584, -0.6905015707015991], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.7118463516235352, -0.6798292398452759], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002832.pt"}
{"idx": 2834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TvMAUk3Etjw.mp4", "duration": 187.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does a player use a piston mechanism to hide inside a block?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a player use a piston mechanism to hide inside a block?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [-0.48772674798965454, -0.4450373649597168], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.4770544171333313, -0.4450373649597168], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002834.pt"}
{"idx": 2836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dNJdJIwCF_Y.mp4", "duration": 100.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone sharpening a knife?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone sharpening a knife?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.9206611514091492, -0.8809917569160461], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9206611514091492, -0.8809917569160461], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002836.pt"}
{"idx": 2838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dNJdJIwCF_Y.mp4", "duration": 100.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where someone is squeezing a golf ball that has been cut in half.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where someone is squeezing a golf ball that has been cut in half.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [0.031404972076416016, 0.11074376106262207], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [0.031404972076416016, 0.11074376106262207], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002838.pt"}
{"idx": 2840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dNJdJIwCF_Y.mp4", "duration": 100.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for dipping a poker chip into the green paste?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for dipping a poker chip into the green paste?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 94.0], "base_se_norm": [0.7851239442825317, 0.8644627332687378], "gt_se_raw": [90.0, 94.0], "gt_se_norm": [0.7851239442825317, 0.8644627332687378], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002840.pt"}
{"idx": 2842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LdmGQg6T29A.mp4", "duration": 22.389033333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl covers her mouth with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl covers her mouth with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.3746938705444336, 0.2506122589111328], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.3746938705444336, 0.16128277778625488], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002842.pt"}
{"idx": 2844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LdmGQg6T29A.mp4", "duration": 22.389033333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman in the purple shirt is holding the girl's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman in the purple shirt is holding the girl's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.3399416208267212], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [0.2506122589111328, 0.6972594261169434], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002844.pt"}
{"idx": 2846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tqpPFT-F-bs.mp4", "duration": 127.5274, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black shirt announce the fourth slap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black shirt announce the fourth slap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 32.0], "base_se_norm": [-0.6079273819923401, -0.49814707040786743], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.9059025645256042, -0.8902196884155273], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002846.pt"}
{"idx": 2848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tqpPFT-F-bs.mp4", "duration": 127.5274, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man in the grey v-neck sweater say 'I'm still in love with you'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man in the grey v-neck sweater say 'I'm still in love with you'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 56.0], "base_se_norm": [-0.153123140335083, -0.12175732851028442], "gt_se_raw": [58.0, 59.0], "gt_se_norm": [-0.09039151668548584, -0.07470864057540894], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002848.pt"}
{"idx": 2850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tqpPFT-F-bs.mp4", "duration": 127.5274, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the man in the suit is tied to a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the man in the suit is tied to a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 103.0], "base_se_norm": [0.36441266536712646, 0.6153391599655151], "gt_se_raw": [86.0, 88.0], "gt_se_norm": [0.34872984886169434, 0.38009560108184814], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002850.pt"}
{"idx": 2852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9cW0Hlj_P-A.mp4", "duration": 189.68285828984344, "prompt_text": "Please find the visual event described by the sentence 'A woman is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 52.0], "base_se_norm": [-0.6415069103240967, -0.45171642303466797], "gt_se_raw": [34.0, 52.0], "gt_se_norm": [-0.6415069103240967, -0.45171642303466797], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002852.pt"}
{"idx": 2854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9cW0Hlj_P-A.mp4", "duration": 189.68285828984344, "prompt_text": "Please find the visual event described by the sentence 'Show the man with glasses sitting on the couch and talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man with glasses sitting on the couch and talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 132.0], "base_se_norm": [0.11765503883361816, 0.3917968273162842], "gt_se_raw": [106.0, 132.0], "gt_se_norm": [0.11765503883361816, 0.3917968273162842], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002854.pt"}
{"idx": 2856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3heXk6Oj6hU.mp4", "duration": 75.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the person writing on the notepad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person writing on the notepad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 38.0], "base_se_norm": [-0.9734630584716797, 0.008403301239013672], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.6550198793411255], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002856.pt"}
{"idx": 2858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3heXk6Oj6hU.mp4", "duration": 75.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman says 'music'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman says 'music'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 63.0], "base_se_norm": [0.6187527179718018, 0.6718266010284424], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.5754091143608093, -0.5223352909088135], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002858.pt"}
{"idx": 2860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3heXk6Oj6hU.mp4", "duration": 75.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman say 'Number four, PE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman say 'Number four, PE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [0.11455106735229492, 0.22069871425628662], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [0.5922157764434814, 0.6718266010284424], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002860.pt"}
{"idx": 2862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QL3H7CUJMDU.mp4", "duration": 160.84, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the baby is shot out of a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the baby is shot out of a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 37.0], "base_se_norm": [-0.5647848844528198, -0.5399154424667358], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.5647848844528198, -0.5399154424667358], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002862.pt"}
{"idx": 2864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QL3H7CUJMDU.mp4", "duration": 160.84, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is eaten by the whale.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is eaten by the whale.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 108.0], "base_se_norm": [0.3180800676345825, 0.3429495096206665], "gt_se_raw": [107.0, 108.0], "gt_se_norm": [0.3305147886276245, 0.3429495096206665], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002864.pt"}
{"idx": 2866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ISQfQ6XmwQo.mp4", "duration": 31.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see an aerial view of the university building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see an aerial view of the university building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.6838777661323547, -0.43097996711730957], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.6838777661323547, -0.49420440196990967], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002866.pt"}
{"idx": 2868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ISQfQ6XmwQo.mp4", "duration": 31.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are people rowing a boat on the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people rowing a boat on the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.8103266954421997, -0.6838777661323547], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8103266954421997, -0.6838777661323547], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002868.pt"}
{"idx": 2870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ISQfQ6XmwQo.mp4", "duration": 31.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the inside of an old-fashioned room with chairs and a fireplace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the inside of an old-fashioned room with chairs and a fireplace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.3045310974121094, -0.1780821681022644], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.3677555322647095, -0.24130666255950928], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002870.pt"}
{"idx": 2872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/75Qg6ldZNbA.mp4", "duration": 198.03116666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the step where the top corners are folded towards the center crease, forming a triangle shape on top.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the step where the top corners are folded towards the center crease, forming a triangle shape on top.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 71.0], "base_se_norm": [-0.45463132858276367, -0.28294116258621216], "gt_se_raw": [54.0, 77.0], "gt_se_norm": [-0.45463132858276367, -0.22234463691711426], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002872.pt"}
{"idx": 2874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/75Qg6ldZNbA.mp4", "duration": 198.03116666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the paper is opened up and flattened from a triangle into a square.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the paper is opened up and flattened from a triangle into a square.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 128.0], "base_se_norm": [0.14123451709747314, 0.2927258014678955], "gt_se_raw": [95.0, 114.0], "gt_se_norm": [-0.04055511951446533, 0.15133392810821533], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002874.pt"}
{"idx": 2876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rbNB0jqMv7s.mp4", "duration": 54.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'When does the instructor in the red suit hold the person in the blue suit upside down by their legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the instructor in the red suit hold the person in the blue suit upside down by their legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 23.0], "base_se_norm": [-0.9631901979446411, -0.15337419509887695], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.33742332458496094, -0.19018405675888062], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002876.pt"}
{"idx": 2878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rbNB0jqMv7s.mp4", "duration": 54.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point does the instructor in the red suit start dancing alone in the tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the instructor in the red suit start dancing alone in the tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 48.0], "base_se_norm": [0.17791414260864258, 0.7668712139129639], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [0.17791414260864258, 0.28834354877471924], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002878.pt"}
{"idx": 2880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rbNB0jqMv7s.mp4", "duration": 54.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the person in the blue suit, wearing a helmet with ears, is lifted up by the instructor and screams with excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the person in the blue suit, wearing a helmet with ears, is lifted up by the instructor and screams with excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.5950920581817627, -0.4478527307510376], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.558282196521759, -0.4478527307510376], "start_token_span": [84, 88], "end_token_span": [90, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002880.pt"}
{"idx": 2882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/StAF3NSro-w.mp4", "duration": 169.3, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy pick up the orange, lumpy ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy pick up the orange, lumpy ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 62.0], "base_se_norm": [-0.4211459159851074, -0.2675723433494568], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [-0.3030124306678772, -0.2793857455253601], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002882.pt"}
{"idx": 2884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/StAF3NSro-w.mp4", "duration": 169.3, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the man places a brown clay-like object and a black square on the shelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the man places a brown clay-like object and a black square on the shelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 123.0], "base_se_norm": [0.22858822345733643, 0.4530419111251831], "gt_se_raw": [104.0, 106.0], "gt_se_norm": [0.22858822345733643, 0.25221502780914307], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002884.pt"}
{"idx": 2886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvfxpJn7VRM.mp4", "duration": 53.59520833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a figurine of a cat standing on a pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a figurine of a cat standing on a pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8507329225540161], "gt_se_raw": [1.0, 18.0], "gt_se_norm": [-0.9626832008361816, -0.3282981514930725], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002886.pt"}
{"idx": 2888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvfxpJn7VRM.mp4", "duration": 53.59520833333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man give a sinister smile to the camera before attacking the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man give a sinister smile to the camera before attacking the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.6268323063850403, -0.4775652289390564], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.5895155668258667, -0.4775652289390564], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002888.pt"}
{"idx": 2890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvfxpJn7VRM.mp4", "duration": 53.59520833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man eats a piece of the smashed pie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man eats a piece of the smashed pie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.10439753532409668, 0.007552742958068848], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.06708073616027832, 0.007552742958068848], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002890.pt"}
{"idx": 2892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WnW-fl0mNwo.mp4", "duration": 79.81306666666667, "prompt_text": "Please find the visual event described by the sentence 'At what moment do the Yankees players storm the field to celebrate their victory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment do the Yankees players storm the field to celebrate their victory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.8245900869369507, -0.6742388010025024], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.8245900869369507, -0.724355936050415], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002892.pt"}
{"idx": 2894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WnW-fl0mNwo.mp4", "duration": 79.81306666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the wide-angle shot of the celebration with the stadium lights shining brightly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the wide-angle shot of the celebration with the stadium lights shining brightly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 35.0], "base_se_norm": [-0.4487118124961853, -0.12295061349868774], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.4487118124961853, -0.3735361695289612], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002894.pt"}
{"idx": 2896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34zJjqZoX4E.mp4", "duration": 101.39295833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the man walking tiredly into the room with a briefcase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man walking tiredly into the room with a briefcase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.8421981334686279, -0.743571937084198], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.822472870349884, -0.7238466739654541], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002896.pt"}
{"idx": 2898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34zJjqZoX4E.mp4", "duration": 101.39295833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the man imagines becoming the number one YouTuber.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the man imagines becoming the number one YouTuber.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 51.0], "base_se_norm": [-0.4279681444168091, 0.005987048149108887], "gt_se_raw": [41.0, 47.0], "gt_se_norm": [-0.19126534461975098, -0.07291394472122192], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002898.pt"}
{"idx": 2900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34zJjqZoX4E.mp4", "duration": 101.39295833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man's face show clear panic and fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man's face show clear panic and fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 77.0], "base_se_norm": [0.43994224071502686, 0.5188431739807129], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [0.43994224071502686, 0.538568377494812], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002900.pt"}
{"idx": 2902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mG4Y2Snygfk.mp4", "duration": 59.02563333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the stick figure is driving a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the stick figure is driving a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.05039107799530029, 0.15204179286956787], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [0.05039107799530029, 0.15204179286956787], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002902.pt"}
{"idx": 2904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mG4Y2Snygfk.mp4", "duration": 59.02563333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range where the drawing transforms the wheelchair symbol into the word 'CAN!'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range where the drawing transforms the wheelchair symbol into the word 'CAN!'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 58.0], "base_se_norm": [0.6941791772842407, 0.9652478694915771], "gt_se_raw": [50.0, 58.0], "gt_se_norm": [0.6941791772842407, 0.9652478694915771], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002904.pt"}
{"idx": 2906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eXDSa-7VWpI.mp4", "duration": 117.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a panoramic view of the village filled with dragons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a panoramic view of the village filled with dragons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 18.0], "base_se_norm": [-0.8468953967094421, -0.6937907934188843], "gt_se_raw": [9.0, 18.0], "gt_se_norm": [-0.8468953967094421, -0.6937907934188843], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002906.pt"}
{"idx": 2908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eXDSa-7VWpI.mp4", "duration": 117.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the blonde girl punches the boy in the arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the blonde girl punches the boy in the arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.08137226104736328, -0.030337393283843994], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.14941877126693726, -0.11539548635482788], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002908.pt"}
{"idx": 2910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eXDSa-7VWpI.mp4", "duration": 117.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the black dragon pounces on a person in the crowd.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the black dragon pounces on a person in the crowd.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 69.0], "base_se_norm": [0.12276721000671387, 0.17380213737487793], "gt_se_raw": [67.0, 69.0], "gt_se_norm": [0.13977885246276855, 0.17380213737487793], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002910.pt"}
{"idx": 2912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sEqSIwf-0Lg.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the water is poured into the filter for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the water is poured into the filter for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 27.0], "base_se_norm": [-0.4752624034881592, -0.32533735036849976], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.4752624034881592, -0.3503248691558838], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002912.pt"}
{"idx": 2914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sEqSIwf-0Lg.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'At what point is the filter being rinsed with boiled water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the filter being rinsed with boiled water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.7501249313354492, -0.6751624345779419], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.7501249313354492, -0.6751624345779419], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002914.pt"}
{"idx": 2916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KY-4VyUY8hk.mp4", "duration": 151.88506666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the butter and brown sugar mixed together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the butter and brown sugar mixed together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [-0.6708037257194519, -0.5391252040863037], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.6708037257194519, -0.5391252040863037], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002916.pt"}
{"idx": 2918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KY-4VyUY8hk.mp4", "duration": 151.88506666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is flour incorporated into the dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is flour incorporated into the dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 76.0], "base_se_norm": [-0.2230967879295349, 0.0007567405700683594], "gt_se_raw": [63.0, 76.0], "gt_se_norm": [-0.17042535543441772, 0.0007567405700683594], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002918.pt"}
{"idx": 2920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KY-4VyUY8hk.mp4", "duration": 151.88506666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the cookie dough balls being rolled and placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the cookie dough balls being rolled and placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 115.0], "base_se_norm": [0.17193877696990967, 0.5143028497695923], "gt_se_raw": [89.0, 110.0], "gt_se_norm": [0.17193877696990967, 0.44846367835998535], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002920.pt"}
{"idx": 2922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KauXf6nihPY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the phone is displayed on a blue background with white rays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the phone is displayed on a blue background with white rays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 12.0], "base_se_norm": [-0.5414421558380127, -0.2139008641242981], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.5414421558380127, -0.2139008641242981], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002922.pt"}
{"idx": 2924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KauXf6nihPY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the female instructor is giving a lecture on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the female instructor is giving a lecture on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.5066900253295898, 0.7032147645950317], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.5066900253295898, 0.6377065181732178], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002924.pt"}
{"idx": 2926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n2Y6in_CZ6A.mp4", "duration": 173.0397063730397, "prompt_text": "Please find the visual event described by the sentence 'Show the exterior view of the California Science Center building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the exterior view of the California Science Center building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 29.0], "base_se_norm": [-0.7803972363471985, -0.6648167967796326], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.7803972363471985, -0.7457231283187866], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002926.pt"}
{"idx": 2928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n2Y6in_CZ6A.mp4", "duration": 173.0397063730397, "prompt_text": "Please find the visual event described by the sentence 'When are two women cleaning the floor and counters in a kitchen-like area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women cleaning the floor and counters in a kitchen-like area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 96.0], "base_se_norm": [0.05178177356719971, 0.10957193374633789], "gt_se_raw": [91.0, 96.0], "gt_se_norm": [0.05178177356719971, 0.10957193374633789], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002928.pt"}
{"idx": 2930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n2Y6in_CZ6A.mp4", "duration": 173.0397063730397, "prompt_text": "Please find the visual event described by the sentence 'Where can we see an octopus trying to open a container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see an octopus trying to open a container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 149.0], "base_se_norm": [0.5834516286849976, 0.7221481800079346], "gt_se_raw": [138.0, 150.0], "gt_se_norm": [0.5950096845626831, 0.7337061166763306], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002930.pt"}
{"idx": 2932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F85cuRuQjLE.mp4", "duration": 193.72, "prompt_text": "Please find the visual event described by the sentence 'At what point does the vehicle start pulling the cow with a rope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the vehicle start pulling the cow with a rope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 169.0], "base_se_norm": [0.42473673820495605, 0.744786262512207], "gt_se_raw": [138.0, 148.0], "gt_se_norm": [0.42473673820495605, 0.5279785394668579], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002932.pt"}
{"idx": 2934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F85cuRuQjLE.mp4", "duration": 193.72, "prompt_text": "Please find the visual event described by the sentence 'When does the camera show a close-up of the medicine bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera show a close-up of the medicine bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 74.0], "base_se_norm": [-0.287631630897522, -0.23601073026657104], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [-0.287631630897522, -0.23601073026657104], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002934.pt"}
{"idx": 2936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3CJl_S7uzqU.mp4", "duration": 124.49103333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man is clapping his hands with excitement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man is clapping his hands with excitement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9518037438392639, -0.9036074876785278], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.9518037438392639, -0.9036074876785278], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002936.pt"}
{"idx": 2938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3CJl_S7uzqU.mp4", "duration": 124.49103333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is someone shown operating a smartphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is someone shown operating a smartphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 64.0], "base_se_norm": [-0.48590677976608276, 0.028186440467834473], "gt_se_raw": [31.0, 36.0], "gt_se_norm": [-0.5019721984863281, -0.4216451048851013], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002938.pt"}
{"idx": 2940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3CJl_S7uzqU.mp4", "duration": 124.49103333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when both the man and the woman give a thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when both the man and the woman give a thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 120.0], "base_se_norm": [0.8957188129425049, 0.9278496503829956], "gt_se_raw": [118.0, 120.0], "gt_se_norm": [0.8957188129425049, 0.9278496503829956], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002940.pt"}
{"idx": 2942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zOVoL1-ZaQQ.mp4", "duration": 145.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip from the 1932 movie 'Freaks'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip from the 1932 movie 'Freaks'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 46.0], "base_se_norm": [-0.43448275327682495, -0.365517258644104], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.43448275327682495, -0.365517258644104], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002942.pt"}
{"idx": 2944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zOVoL1-ZaQQ.mp4", "duration": 145.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the movie 'Charly' is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the movie 'Charly' is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 92.0], "base_se_norm": [0.18620693683624268, 0.268965482711792], "gt_se_raw": [87.0, 91.0], "gt_se_norm": [0.20000004768371582, 0.25517237186431885], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002944.pt"}
{"idx": 2946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/exKximzPvAs.mp4", "duration": 101.51808333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the part of the book about decorating with balloons and ribbon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part of the book about decorating with balloons and ribbon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.6256824731826782, -0.5468787550926208], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.6256824731826782, -0.5468787550926208], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002946.pt"}
{"idx": 2948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/exKximzPvAs.mp4", "duration": 101.51808333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the page where 'Tania wore a mask'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the page where 'Tania wore a mask'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [-0.15286028385162354, -0.07405656576156616], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.15286028385162354, -0.07405656576156616], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002948.pt"}
{"idx": 2950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/exKximzPvAs.mp4", "duration": 101.51808333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point is a boy shown drilling into a piece of wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a boy shown drilling into a piece of wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 88.0], "base_se_norm": [0.6745785474777222, 0.733681321144104], "gt_se_raw": [84.0, 88.0], "gt_se_norm": [0.6548776626586914, 0.733681321144104], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002950.pt"}
{"idx": 2952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y-Z7LObUlwA.mp4", "duration": 155.3552, "prompt_text": "Please find the visual event described by the sentence 'Show me when the camera enters the main lobby and reception area.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the camera enters the main lobby and reception area.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 32.0], "base_se_norm": [-0.716778039932251, -0.588040828704834], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.7039043307304382, -0.6524094343185425], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002952.pt"}
{"idx": 2954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y-Z7LObUlwA.mp4", "duration": 155.3552, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the camera provides a panoramic view of the testing room cubicles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the camera provides a panoramic view of the testing room cubicles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 71.0], "base_se_norm": [-0.1889553666114807, -0.08596557378768921], "gt_se_raw": [65.0, 71.0], "gt_se_norm": [-0.16320788860321045, -0.08596557378768921], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002954.pt"}
{"idx": 2956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K7bNhYr3ves.mp4", "duration": 84.61786666666667, "prompt_text": "Please find the visual event described by the sentence 'When do Marlin and Dory see Nemo lying motionless on the seabed inside the net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do Marlin and Dory see Nemo lying motionless on the seabed inside the net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.8345503211021423, -0.6454649567604065], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.5981936454772949, -0.48001527786254883], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002956.pt"}
{"idx": 2958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K7bNhYr3ves.mp4", "duration": 84.61786666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when Nemo tells his dad, 'I don't hate you.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when Nemo tells his dad, 'I don't hate you.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.4890472888946533, 0.6072255373001099], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [-0.14911586046218872, -0.10184454917907715], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002958.pt"}
{"idx": 2960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K7bNhYr3ves.mp4", "duration": 84.61786666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the fish being lifted out of the water in a large green net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the fish being lifted out of the water in a large green net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.8581860065460205, -0.7163719534873962], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8345503211021423, -0.787278950214386], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002960.pt"}
{"idx": 2962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpNiqsM81K0.mp4", "duration": 41.37466666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person on the right start tapping the other person's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person on the right start tapping the other person's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 34.0], "base_se_norm": [-0.5166124105453491, 0.6435177326202393], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.5166124105453491, -0.4199349284172058], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002962.pt"}
{"idx": 2964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpNiqsM81K0.mp4", "duration": 41.37466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the two people begin to separate their interlocked hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the two people begin to separate their interlocked hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [0.5951789617538452, 0.933550238609314], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [0.35348522663116455, 0.49850142002105713], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002964.pt"}
{"idx": 2966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/igOTq9Z7bm0.mp4", "duration": 86.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the child's hand tapping the adult's leg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the child's hand tapping the adult's leg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.7926068305969238], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.884781539440155], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002966.pt"}
{"idx": 2968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/igOTq9Z7bm0.mp4", "duration": 86.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the white shirt start touching the boy's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the white shirt start touching the boy's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 23.0], "base_se_norm": [-0.6313009858131409, -0.4699952006340027], "gt_se_raw": [16.0, 23.0], "gt_se_norm": [-0.6313009858131409, -0.4699952006340027], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002968.pt"}
{"idx": 2970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/igOTq9Z7bm0.mp4", "duration": 86.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy say 'Game Over!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy say 'Game Over!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 82.0], "base_se_norm": [0.8204513788223267, 0.8895823955535889], "gt_se_raw": [79.0, 81.0], "gt_se_norm": [0.8204513788223267, 0.8665386438369751], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002970.pt"}
{"idx": 2972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-MuOaopJY64.mp4", "duration": 9.509500000000001, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame in which the child is running or skipping ahead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame in which the child is running or skipping ahead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.369051992893219, 0.4722120761871338], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.369051992893219], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002972.pt"}
{"idx": 2974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-MuOaopJY64.mp4", "duration": 9.509500000000001, "prompt_text": "Please find the visual event described by the sentence 'During which period are their reflections clearly visible on the polished floor as they walk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which period are their reflections clearly visible on the polished floor as they walk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.8928440809249878], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, 0.8928440809249878], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002974.pt"}
{"idx": 2976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-_kGBuX1Qy0.mp4", "duration": 11.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl in the pink jacket laughing out loud?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl in the pink jacket laughing out loud?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.45454543828964233, 0.09090912342071533], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.45454543828964233, -0.09090906381607056], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002976.pt"}
{"idx": 2978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-_kGBuX1Qy0.mp4", "duration": 11.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman with red hair holds up a red bubble wand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman with red hair holds up a red bubble wand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.8181818723678589], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.45454543828964233], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002978.pt"}
{"idx": 2980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-_kGBuX1Qy0.mp4", "duration": 11.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman says 'She is laughing'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman says 'She is laughing'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [0.27272725105285645, 0.6363636255264282], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [0.27272725105285645, 0.6363636255264282], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002980.pt"}
{"idx": 2982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O9T25SMyz3A.mp4", "duration": 171.80496666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man dials 911 on a wall-mounted phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man dials 911 on a wall-mounted phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [-0.35973912477493286, -0.3248157501220703], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [-0.3480979800224304, -0.3248157501220703], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002982.pt"}
{"idx": 2984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O9T25SMyz3A.mp4", "duration": 171.80496666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the correct CPR method of 'CAB' (Compressions, Airway, Breathing) explained with on-screen text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the correct CPR method of 'CAB' (Compressions, Airway, Breathing) explained with on-screen text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 92.0], "base_se_norm": [-0.06871140003204346, 0.07098186016082764], "gt_se_raw": [87.0, 92.0], "gt_se_norm": [0.012776374816894531, 0.07098186016082764], "start_token_span": [83, 87], "end_token_span": [89, 93], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002984.pt"}
{"idx": 2986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkDMIKoZDOs.mp4", "duration": 30.466666666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the person sitting on the grass and putting on their shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person sitting on the grass and putting on their shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.4748358726501465], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.4748358726501465], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002986.pt"}
{"idx": 2988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkDMIKoZDOs.mp4", "duration": 30.466666666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the runner gets into the starting blocks position.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the runner gets into the starting blocks position.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.015317320823669434, 0.24726474285125732], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.015317320823669434, 0.3129103183746338], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002988.pt"}
{"idx": 2990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkDMIKoZDOs.mp4", "duration": 30.466666666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the 'JUST DO IT.' slogan appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the 'JUST DO IT.' slogan appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [0.7724288702011108, 0.9037199020385742], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.7724288702011108, 0.9037199020385742], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002990.pt"}
{"idx": 2992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WH50zXiSVPg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the narrator says people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the narrator says people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.12708652019500732, 0.35786545276641846], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002992.pt"}
{"idx": 2994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WH50zXiSVPg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what time is the 84.4% employment rate for non-disabled Americans in 1991 visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the 84.4% employment rate for non-disabled Americans in 1991 visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [85, 88], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002994.pt"}
{"idx": 2996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NQQH5TfyNKc.mp4", "duration": 58.25825825825826, "prompt_text": "Please find the visual event described by the sentence 'When is the first woman speaking in front of the AXIS sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the first woman speaking in front of the AXIS sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.931340217590332, -0.6223711371421814], "gt_se_raw": [2.0, 11.0], "gt_se_norm": [-0.931340217590332, -0.6223711371421814], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002996.pt"}
{"idx": 2998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NQQH5TfyNKc.mp4", "duration": 58.25825825825826, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the black blazer speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the black blazer speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 37.0], "base_se_norm": [0.06422674655914307, 0.2702062129974365], "gt_se_raw": [30.0, 38.0], "gt_se_norm": [0.029896855354309082, 0.3045361042022705], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002998.pt"}
{"idx": 3000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NQQH5TfyNKc.mp4", "duration": 58.25825825825826, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man with a beard and suspenders talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man with a beard and suspenders talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 55.0], "base_se_norm": [0.6135051250457764, 0.8881442546844482], "gt_se_raw": [46.0, 55.0], "gt_se_norm": [0.5791752338409424, 0.8881442546844482], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003000.pt"}
{"idx": 3002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwBxO3289cE.mp4", "duration": 132.93293293293294, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman with short hair and hoop earrings speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman with short hair and hoop earrings speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 44.0], "base_se_norm": [-0.458373486995697, -0.33801209926605225], "gt_se_raw": [35.0, 44.0], "gt_se_norm": [-0.4734187126159668, -0.33801209926605225], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003002.pt"}
{"idx": 3004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwBxO3289cE.mp4", "duration": 132.93293293293294, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the text 'TRANSCRIPTS' is displayed on a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the text 'TRANSCRIPTS' is displayed on a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 95.0], "base_se_norm": [0.33902108669281006, 0.42929208278656006], "gt_se_raw": [90.0, 95.0], "gt_se_norm": [0.35406625270843506, 0.42929208278656006], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003004.pt"}
{"idx": 3006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q5-KoPvZDIU.mp4", "duration": 120.08663333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a close-up of a dragonfly's red and white compound eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a close-up of a dragonfly's red and white compound eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [0.08255183696746826, 0.19913434982299805], "gt_se_raw": [65.0, 72.0], "gt_se_norm": [0.08255183696746826, 0.19913434982299805], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003006.pt"}
{"idx": 3008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q5-KoPvZDIU.mp4", "duration": 120.08663333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a dragonfly takes off from a twig in slow motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a dragonfly takes off from a twig in slow motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [-0.2505410313606262, -0.20057713985443115], "gt_se_raw": [35.0, 48.0], "gt_se_norm": [-0.41708749532699585, -0.20057713985443115], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003008.pt"}
{"idx": 3010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q5-KoPvZDIU.mp4", "duration": 120.08663333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a red dragonfly is shown resting on a twig above the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a red dragonfly is shown resting on a twig above the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 24.0], "base_se_norm": [-0.8167989253997803, -0.6002885699272156], "gt_se_raw": [11.0, 23.0], "gt_se_norm": [-0.8167989253997803, -0.6169432401657104], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003010.pt"}
{"idx": 3012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8ISzf2pryI.mp4", "duration": 138.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the labrador retriever starts to show its teeth guiltily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the labrador retriever starts to show its teeth guiltily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [0.11594200134277344, 0.20289850234985352], "gt_se_raw": [76.0, 82.0], "gt_se_norm": [0.10144925117492676, 0.18840575218200684], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003012.pt"}
{"idx": 3014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8ISzf2pryI.mp4", "duration": 138.0, "prompt_text": "Please find the visual event described by the sentence 'Where in the video is the guilty dog sent to the kennel and walks away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video is the guilty dog sent to the kennel and walks away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 120.0], "base_se_norm": [0.6376811265945435, 0.7391303777694702], "gt_se_raw": [112.0, 120.0], "gt_se_norm": [0.6231883764266968, 0.7391303777694702], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003014.pt"}
{"idx": 3016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W2Cxs_p2-oE.mp4", "duration": 127.127, "prompt_text": "Please find the visual event described by the sentence 'When is the princess checking her pink phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the princess checking her pink phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.622424840927124, -0.5752279162406921], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.622424840927124, -0.5752279162406921], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003016.pt"}
{"idx": 3018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W2Cxs_p2-oE.mp4", "duration": 127.127, "prompt_text": "Please find the visual event described by the sentence 'Show the part where Cinderella's dress is transformed into a plain t-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where Cinderella's dress is transformed into a plain t-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 120.0], "base_se_norm": [0.7305529117584229, 0.8878759145736694], "gt_se_raw": [109.0, 112.0], "gt_se_norm": [0.7148206233978271, 0.7620174884796143], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003018.pt"}
{"idx": 3020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W2Cxs_p2-oE.mp4", "duration": 127.127, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when Cinderella shouts for help.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when Cinderella shouts for help.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 50.0], "base_se_norm": [-0.27631425857543945, -0.21338504552841187], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.5594956278800964, -0.48083412647247314], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003020.pt"}
{"idx": 3022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B43fpt5MYxc.mp4", "duration": 151.70833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man in a blue patterned shirt is filling out a form at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man in a blue patterned shirt is filling out a form at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.44630593061447144, -0.39357316493988037], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [-0.44630593061447144, -0.39357316493988037], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003022.pt"}
{"idx": 3024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B43fpt5MYxc.mp4", "duration": 151.70833333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where a health and safety guidelines sign is shown on a glass door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where a health and safety guidelines sign is shown on a glass door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 59.0], "base_se_norm": [-0.3012908101081848, -0.2221916913986206], "gt_se_raw": [53.0, 59.0], "gt_se_norm": [-0.3012908101081848, -0.2221916913986206], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003024.pt"}
{"idx": 3026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4FE1CLS7i3k.mp4", "duration": 312.8542083333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man wearing a red shirt and holding a sign that says 'I use sign language'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man wearing a red shirt and holding a sign that says 'I use sign language'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 118.0], "base_se_norm": [-0.2904043197631836, -0.24565505981445312], "gt_se_raw": [111.0, 114.0], "gt_se_norm": [-0.2904043197631836, -0.27122604846954346], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003026.pt"}
{"idx": 3028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4FE1CLS7i3k.mp4", "duration": 312.8542083333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman hold up the sign that says 'I failed math 3 times'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman hold up the sign that says 'I failed math 3 times'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 144.0], "base_se_norm": [-0.11140722036361694, -0.07944345474243164], "gt_se_raw": [138.0, 142.0], "gt_se_norm": [-0.11779999732971191, -0.0922289490699768], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003028.pt"}
{"idx": 3030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4FE1CLS7i3k.mp4", "duration": 312.8542083333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man wearing a cap and holding an orange sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man wearing a cap and holding an orange sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 110.0], "base_se_norm": [-0.3159753680229187, -0.29679709672927856], "gt_se_raw": [107.0, 110.0], "gt_se_norm": [-0.3159753680229187, -0.29679709672927856], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003030.pt"}
{"idx": 3032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bWy7nCH3zmQ.mp4", "duration": 116.34956666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman points to the 'book, story' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman points to the 'book, story' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.5358814001083374, -0.48431265354156494], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.5358814001083374, -0.48431265354156494], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003032.pt"}
{"idx": 3034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bWy7nCH3zmQ.mp4", "duration": 116.34956666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is the smartphone being shown to the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the smartphone being shown to the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.19208979606628418, -0.0373835563659668], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.19208979606628418, -0.14052104949951172], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003034.pt"}
{"idx": 3036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ua6wauucdOo.mp4", "duration": 151.1092888888889, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a 3D model of the cricket is being manipulated on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a 3D model of the cricket is being manipulated on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 80.0], "base_se_norm": [-0.020576417446136475, 0.05883634090423584], "gt_se_raw": [74.0, 81.0], "gt_se_norm": [-0.020576417446136475, 0.07207179069519043], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003036.pt"}
{"idx": 3038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ua6wauucdOo.mp4", "duration": 151.1092888888889, "prompt_text": "Please find the visual event described by the sentence 'When does the student in the grey shirt drop the cricket into the enclosure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the student in the grey shirt drop the cricket into the enclosure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.351462721824646, -0.3117563724517822], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.3382272720336914, -0.3117563724517822], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003038.pt"}
{"idx": 3040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ua6wauucdOo.mp4", "duration": 151.1092888888889, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when Professor Rajat Mittal introduces himself and his title appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when Professor Rajat Mittal introduces himself and his title appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 45.0], "base_se_norm": [-0.48381727933883667, -0.40440458059310913], "gt_se_raw": [35.0, 42.0], "gt_se_norm": [-0.5367591381072998, -0.4441109299659729], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003040.pt"}
{"idx": 3042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MxNfvh7vaSs.mp4", "duration": 149.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the instructions for the 'find the odd one out' game are given.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the instructions for the 'find the odd one out' game are given.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 59.0], "base_se_norm": [-0.5318769812583923, -0.21087831258773804], "gt_se_raw": [45.0, 75.0], "gt_se_norm": [-0.39812755584716797, 0.0031207799911499023], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003042.pt"}
{"idx": 3044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MxNfvh7vaSs.mp4", "duration": 149.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the user successfully identify the 'imposter' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the user successfully identify the 'imposter' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.49175214767456055, -0.45162731409072876], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [0.23049485683441162, 0.27061963081359863], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003044.pt"}
{"idx": 3046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DiZEribdeTw.mp4", "duration": 51.176125, "prompt_text": "Please find the visual event described by the sentence 'When does the shadow of a person carrying an axe walk across the screen from left to right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the shadow of a person carrying an axe walk across the screen from left to right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.9218385219573975, -0.7655156850814819], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9218385219573975, -0.8436771035194397], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003046.pt"}
{"idx": 3048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DiZEribdeTw.mp4", "duration": 51.176125, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man crouches down in front of the log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man crouches down in front of the log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.6091927289962769, -0.45286989212036133], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.5701120495796204, -0.45286989212036133], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003048.pt"}
{"idx": 3050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DiZEribdeTw.mp4", "duration": 51.176125, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man picks up the two split pieces of wood and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man picks up the two split pieces of wood and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 27.0], "base_se_norm": [-0.17930477857589722, 0.055179595947265625], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.022981882095336914, 0.055179595947265625], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003050.pt"}
{"idx": 3052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QFMMD4JwTwM.mp4", "duration": 15.482133333333334, "prompt_text": "Please find the visual event described by the sentence 'A player in an orange jersey skates with the puck towards the goal. When is this happening?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A player in an orange jersey skates with the puck towards the goal. When is this happening?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 12.0], "base_se_norm": [-0.4832753539085388, 0.5501739978790283], "gt_se_raw": [6.0, 11.0], "gt_se_norm": [-0.22491300106048584, 0.4209928512573242], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003052.pt"}
{"idx": 3054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QFMMD4JwTwM.mp4", "duration": 15.482133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a goal is scored.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a goal is scored.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.4209928512573242, 0.6793551445007324], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [0.5501739978790283, 0.6793551445007324], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003054.pt"}
{"idx": 3056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txqiwrbYGrs.mp4", "duration": 118.9, "prompt_text": "Please find the visual event described by the sentence 'When does the boy ask 'Is this real life?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy ask 'Is this real life?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.19259881973266602, -0.10849452018737793], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.6972245573997498, -0.6635828614234924], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003056.pt"}
{"idx": 3058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txqiwrbYGrs.mp4", "duration": 118.9, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boy starts to scream loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boy starts to scream loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [-0.04121112823486328, 0.00925147533416748], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [-0.04121112823486328, -0.007569372653961182], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003058.pt"}
{"idx": 3060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txqiwrbYGrs.mp4", "duration": 118.9, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the boy asks if this feeling will last forever.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the boy asks if this feeling will last forever.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 72.0], "base_se_norm": [0.1269974708557129, 0.21110177040100098], "gt_se_raw": [105.0, 107.0], "gt_se_norm": [0.7661900520324707, 0.799831748008728], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003060.pt"}
{"idx": 3062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/twKxRnoGxoM.mp4", "duration": 65.73233333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man holding the sign in the hospital bed for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man holding the sign in the hospital bed for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 14.0], "base_se_norm": [-0.817441463470459, -0.5740300416946411], "gt_se_raw": [6.0, 14.0], "gt_se_norm": [-0.817441463470459, -0.5740300416946411], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003062.pt"}
{"idx": 3064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/twKxRnoGxoM.mp4", "duration": 65.73233333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is waving his hand to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is waving his hand to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 42.0], "base_se_norm": [0.06492495536804199, 0.2779099941253662], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [0.06492495536804199, 0.247483491897583], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003064.pt"}
{"idx": 3066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vy6qB2FLbvE.mp4", "duration": 104.31253333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person is mixing the ingredients for the meatballs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person is mixing the ingredients for the meatballs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.6357101202011108, -0.5398443937301636], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.6357101202011108, -0.5973638296127319], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003066.pt"}
{"idx": 3068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vy6qB2FLbvE.mp4", "duration": 104.31253333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where someone is stirring the cheese and herb mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where someone is stirring the cheese and herb mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.3481128215789795, -0.2714202404022217], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.3481128215789795, -0.29059338569641113], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003068.pt"}
{"idx": 3070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vy6qB2FLbvE.mp4", "duration": 104.31253333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a knife and fork are cutting into a finished stuffed shell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a knife and fork are cutting into a finished stuffed shell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [0.4763326644897461, 0.5530253648757935], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.49550580978393555, 0.5530253648757935], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003070.pt"}
{"idx": 3072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lyoaLway15g.mp4", "duration": 199.29166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the girl is happily browsing through records in the store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the girl is happily browsing through records in the store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 52.0], "base_se_norm": [-0.6487560272216797, -0.478151798248291], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.628684937953949, -0.5885427594184875], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003072.pt"}
{"idx": 3074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lyoaLway15g.mp4", "duration": 199.29166666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl, now dressed up, get off the bus and leave her headphones behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl, now dressed up, get off the bus and leave her headphones behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 109.0], "base_se_norm": [0.02362525463104248, 0.09387409687042236], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [-0.01651686429977417, 0.02362525463104248], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003074.pt"}
{"idx": 3076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WO73wnquXtM.mp4", "duration": 187.7709111111111, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where two women are being interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where two women are being interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.6272052526473999, -0.5419950485229492], "gt_se_raw": [36.0, 40.0], "gt_se_norm": [-0.6165540218353271, -0.573948860168457], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003076.pt"}
{"idx": 3078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WO73wnquXtM.mp4", "duration": 187.7709111111111, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the white beret hold up two different bowls for the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the white beret hold up two different bowls for the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.22489690780639648, 0.2568507194519043], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [0.22489690780639648, 0.2568507194519043], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003078.pt"}
{"idx": 3080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WO73wnquXtM.mp4", "duration": 187.7709111111111, "prompt_text": "Please find the visual event described by the sentence 'When is the avocado being mashed in a metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the avocado being mashed in a metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 148.0], "base_se_norm": [0.5444352626800537, 0.5763890743255615], "gt_se_raw": [147.0, 148.0], "gt_se_norm": [0.5657378435134888, 0.5763890743255615], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003080.pt"}
{"idx": 3082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S7VYhBwbprE.mp4", "duration": 97.2972, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman gestures with her hands clasped together while asking a question.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman gestures with her hands clasped together while asking a question.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.6299996376037598, -0.5477772951126099], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.6299996376037598, -0.5888885259628296], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003082.pt"}
{"idx": 3084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S7VYhBwbprE.mp4", "duration": 97.2972, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman touch her chest area with her right hand while looking surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman touch her chest area with her right hand while looking surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 62.0], "base_se_norm": [0.15111219882965088, 0.27444565296173096], "gt_se_raw": [56.0, 61.0], "gt_se_norm": [0.15111219882965088, 0.25389015674591064], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003084.pt"}
{"idx": 3086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nTgVwn0ZrXM.mp4", "duration": 39.039, "prompt_text": "Please find the visual event described by the sentence 'When are the mother and child playing pat-a-cake by clapping their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the mother and child playing pat-a-cake by clapping their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 33.0], "base_se_norm": [-1.0, 0.6906169652938843], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, -0.3852301836013794], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003086.pt"}
{"idx": 3088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nTgVwn0ZrXM.mp4", "duration": 39.039, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman pat the baby's legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman pat the baby's legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.07784527540206909, 0.07584714889526367], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.02661442756652832, 0.07584714889526367], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003088.pt"}
{"idx": 3090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nTgVwn0ZrXM.mp4", "duration": 39.039, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the baby is being held and tickled by the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the baby is being held and tickled by the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 38.0], "base_se_norm": [-1.0, 0.9467711448669434], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [0.6906169652938843, 0.8955402374267578], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003090.pt"}
{"idx": 3092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtS1l2NXWX8.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the video explains the two types of tactile indicators.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the video explains the two types of tactile indicators.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.42719393968582153, -0.37264102697372437], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.4408321976661682, -0.37264102697372437], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003092.pt"}
{"idx": 3094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtS1l2NXWX8.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the presenter is standing at a crosswalk in Sydney.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the presenter is standing at a crosswalk in Sydney.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 86.0], "base_se_norm": [-0.05896151065826416, 0.17288851737976074], "gt_se_raw": [70.0, 86.0], "gt_se_norm": [-0.04532325267791748, 0.17288851737976074], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003094.pt"}
{"idx": 3096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nX4mmS4k5Co.mp4", "duration": 39.666666666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the customer ask if the store carries salt and pepper for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the customer ask if the store carries salt and pepper for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.7983193397521973, -0.49579834938049316], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.6974790096282959, -0.5462185144424438], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003096.pt"}
{"idx": 3098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nX4mmS4k5Co.mp4", "duration": 39.666666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point is the cashier holding a fan of money?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the cashier holding a fan of money?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [0.3109242916107178, 0.6134452819824219], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.4621847867965698, 0.6134452819824219], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003098.pt"}
{"idx": 3100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nX4mmS4k5Co.mp4", "duration": 39.666666666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the character surrender by raising a white flag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character surrender by raising a white flag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 37.0], "base_se_norm": [0.7647057771682739, 0.8655461072921753], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.7647057771682739, 0.8655461072921753], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003100.pt"}
{"idx": 3102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L3MtFGWRXAA.mp4", "duration": 214.75620833333332, "prompt_text": "Please find the visual event described by the sentence 'Show me when the puppy is licking the man's face while he is lying on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the puppy is licking the man's face while he is lying on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [-0.5250428318977356, -0.5064170360565186], "gt_se_raw": [51.0, 52.0], "gt_se_norm": [-0.5250428318977356, -0.5157299637794495], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003102.pt"}
{"idx": 3104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L3MtFGWRXAA.mp4", "duration": 214.75620833333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man take the puppy to the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man take the puppy to the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 135.0], "base_se_norm": [0.21067512035369873, 0.25723958015441895], "gt_se_raw": [130.0, 134.0], "gt_se_norm": [0.21067512035369873, 0.2479267120361328], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003104.pt"}
{"idx": 3106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gLax3zOBN40.mp4", "duration": 108.9088, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the title 'I AM SAM' inside an octagon shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the title 'I AM SAM' inside an octagon shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9265440702438354], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9449080228805542], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003106.pt"}
{"idx": 3108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gLax3zOBN40.mp4", "duration": 108.9088, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment showing a man and a girl eating at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment showing a man and a girl eating at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.13689249753952026, -0.08180052042007446], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.11852854490280151, -0.08180052042007446], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003108.pt"}
{"idx": 3110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HFa9WfIJgtY.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When do we see the process of adding ground coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see the process of adding ground coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.8001998066902161, -0.7252747416496277], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.8001998066902161, -0.7252747416496277], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003110.pt"}
{"idx": 3112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HFa9WfIJgtY.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'At what point do we observe the coffee being stirred for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we observe the coffee being stirred for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [-0.30069929361343384, -0.22577422857284546], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [-0.30069929361343384, -0.22577422857284546], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003112.pt"}
{"idx": 3114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HFa9WfIJgtY.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When is milk being added to the coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is milk being added to the coffee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 58.0], "base_se_norm": [0.3986014127731323, 0.4485514163970947], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [0.3986014127731323, 0.4485514163970947], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003114.pt"}
{"idx": 3116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IkpWQHLY4A4.mp4", "duration": 184.5, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a person wearing a unicorn mask sits in a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a person wearing a unicorn mask sits in a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.7398374080657959, -0.6747967600822449], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.7398374080657959, -0.6856368780136108], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003116.pt"}
{"idx": 3118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IkpWQHLY4A4.mp4", "duration": 184.5, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in a maroon shirt walking away down the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in a maroon shirt walking away down the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 47.0], "base_se_norm": [-0.5989159941673279, -0.49051493406295776], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.5555555820465088, -0.5013550519943237], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003118.pt"}
{"idx": 3120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzJqeyxye_E.mp4", "duration": 91.42466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the sun rises over the horizon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the sun rises over the horizon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.8468684554100037, -0.7156128883361816], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.8249925374984741, -0.7374887466430664], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003120.pt"}
{"idx": 3122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzJqeyxye_E.mp4", "duration": 91.42466666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the total solar eclipse occur, making the sky dark like night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the total solar eclipse occur, making the sky dark like night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [0.24692821502685547, 0.37818384170532227], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.2906801700592041, 0.37818384170532227], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003122.pt"}
{"idx": 3124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzJqeyxye_E.mp4", "duration": 91.42466666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the moon has completely moved past the sun, ending the eclipse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the moon has completely moved past the sun, ending the eclipse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [0.6844469308853149, 0.7938265800476074], "gt_se_raw": [79.0, 82.0], "gt_se_norm": [0.728198766708374, 0.7938265800476074], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003124.pt"}
{"idx": 3126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F1NR-_YqOgE.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the time segment where the speaker mentions the overall trend.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time segment where the speaker mentions the overall trend.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9031662940979004, -0.03166300058364868], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [-0.2253304123878479, -0.1284967064857483], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003126.pt"}
{"idx": 3128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F1NR-_YqOgE.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the narrator starts explaining the data on the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the narrator starts explaining the data on the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.03166300058364868], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.8063325881958008], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003128.pt"}
{"idx": 3130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I82ffq4niLU.mp4", "duration": 75.54213333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman and a child are playing with a green toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman and a child are playing with a green toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [0.05901157855987549, 0.2708139419555664], "gt_se_raw": [40.0, 48.0], "gt_se_norm": [0.05901157855987549, 0.2708139419555664], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003130.pt"}
{"idx": 3132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I82ffq4niLU.mp4", "duration": 75.54213333333334, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a woman and a boy playing with a yellow bowl of colorful toys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a woman and a boy playing with a yellow bowl of colorful toys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 61.0], "base_se_norm": [0.45614099502563477, 0.6149927377700806], "gt_se_raw": [55.0, 61.0], "gt_se_norm": [0.45614099502563477, 0.6149927377700806], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003132.pt"}
{"idx": 3134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I82ffq4niLU.mp4", "duration": 75.54213333333334, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a blue jacket interact with a young person outdoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a blue jacket interact with a young person outdoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 75.0], "base_se_norm": [0.7473691701889038, 0.9856467247009277], "gt_se_raw": [66.0, 75.0], "gt_se_norm": [0.7473691701889038, 0.9856467247009277], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003134.pt"}
{"idx": 3136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oUCSXtTHo5s.mp4", "duration": 167.83433333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two side corners folded down to create the dog's ears?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two side corners folded down to create the dog's ears?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 127.0], "base_se_norm": [0.12015223503112793, 0.5133970975875854], "gt_se_raw": [67.0, 75.0], "gt_se_norm": [-0.20159363746643066, -0.10626155138015747], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003136.pt"}
{"idx": 3138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oUCSXtTHo5s.mp4", "duration": 167.83433333333335, "prompt_text": "Please find the visual event described by the sentence 'When are the eyes and nose drawn onto the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the eyes and nose drawn onto the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 147.0], "base_se_norm": [0.5848962068557739, 0.7517273426055908], "gt_se_raw": [132.0, 147.0], "gt_se_norm": [0.5729796886444092, 0.7517273426055908], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003138.pt"}
{"idx": 3140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hoQ0_smwA3k.mp4", "duration": 69.53613333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the 'Increase Awareness' section displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'Increase Awareness' section displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.5685696601867676, -0.3672354221343994], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.5973316431045532, -0.39599746465682983], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003140.pt"}
{"idx": 3142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hoQ0_smwA3k.mp4", "duration": 69.53613333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video suggest to 'Sit on your hands if it's hard to not touch your face'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video suggest to 'Sit on your hands if it's hard to not touch your face'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 48.0], "base_se_norm": [0.09295690059661865, 0.3805772066116333], "gt_se_raw": [37.0, 47.0], "gt_se_norm": [0.06419491767883301, 0.35181522369384766], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003142.pt"}
{"idx": 3144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hoQ0_smwA3k.mp4", "duration": 69.53613333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the Indiana University logo is shown at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the Indiana University logo is shown at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [0.8695316314697266, 0.955817699432373], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.8695316314697266, 0.955817699432373], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003144.pt"}
{"idx": 3146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F7Ug863S8dQ.mp4", "duration": 163.20470833333331, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a boy in a striped shirt drinks water from a metal bowl on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a boy in a striped shirt drinks water from a metal bowl on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 59.0], "base_se_norm": [-0.30149078369140625, -0.2769816517829895], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.2892361879348755, -0.2647271156311035], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003146.pt"}
{"idx": 3148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F7Ug863S8dQ.mp4", "duration": 163.20470833333331, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy launch a large missile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy launch a large missile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 97.0], "base_se_norm": [0.16418206691741943, 0.1886911392211914], "gt_se_raw": [94.0, 96.0], "gt_se_norm": [0.15192747116088867, 0.1764366626739502], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003148.pt"}
{"idx": 3150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y_w105aWPNY.mp4", "duration": 157.94258373205741, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the green dress enter the shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the green dress enter the shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 12.0], "base_se_norm": [-0.9113602042198181, -0.848046064376831], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.9113602042198181, -0.8607088923454285], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003150.pt"}
{"idx": 3152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y_w105aWPNY.mp4", "duration": 157.94258373205741, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the snow-covered man puts the carrots on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the snow-covered man puts the carrots on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 70.0], "base_se_norm": [-0.15159040689468384, -0.1136019229888916], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.13892757892608643, -0.1136019229888916], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003152.pt"}
{"idx": 3154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y_w105aWPNY.mp4", "duration": 157.94258373205741, "prompt_text": "Please find the visual event described by the sentence 'When is the man thrown out of the shop and into the snow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man thrown out of the shop and into the snow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 132.0], "base_se_norm": [0.5828536748886108, 0.6714935302734375], "gt_se_raw": [126.0, 129.0], "gt_se_norm": [0.595516562461853, 0.6335049867630005], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003154.pt"}
{"idx": 3156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xrhkfADEtMU.mp4", "duration": 147.52237499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does Patrick mistakenly present a pickle instead of a jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Patrick mistakenly present a pickle instead of a jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.9186564087867737, -0.8644273281097412], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.9050991535186768, -0.8644273281097412], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003156.pt"}
{"idx": 3158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xrhkfADEtMU.mp4", "duration": 147.52237499999998, "prompt_text": "Please find the visual event described by the sentence 'When do SpongeBob and Patrick celebrate with balloons and confetti?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do SpongeBob and Patrick celebrate with balloons and confetti?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 81.0], "base_se_norm": [0.0032377243041992188, 0.09813857078552246], "gt_se_raw": [74.0, 82.0], "gt_se_norm": [0.0032377243041992188, 0.11169576644897461], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003158.pt"}
{"idx": 3160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6n5IGnuadhM.mp4", "duration": 92.994, "prompt_text": "Please find the visual event described by the sentence 'When does the blue character balance on the red tightrope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue character balance on the red tightrope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9784932136535645, -0.7204120755195618], "gt_se_raw": [2.0, 10.0], "gt_se_norm": [-0.9569864869117737, -0.7849323749542236], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003160.pt"}
{"idx": 3162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6n5IGnuadhM.mp4", "duration": 92.994, "prompt_text": "Please find the visual event described by the sentence 'A nine-headed character kidnaps the woman and puts her in a carriage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A nine-headed character kidnaps the woman and puts her in a carriage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 65.0], "base_se_norm": [0.20437872409820557, 0.3979395627975464], "gt_se_raw": [57.0, 62.0], "gt_se_norm": [0.2258855104446411, 0.3334193229675293], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003162.pt"}
{"idx": 3164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6n5IGnuadhM.mp4", "duration": 92.994, "prompt_text": "Please find the visual event described by the sentence 'When is the woman shown holding a picture of the blue man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman shown holding a picture of the blue man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 55.0], "base_se_norm": [-0.11822271347045898, 0.18287193775177002], "gt_se_raw": [81.0, 83.0], "gt_se_norm": [0.7420477867126465, 0.7850613594055176], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003164.pt"}
{"idx": 3166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gyktHss8qmU.mp4", "duration": 59.893166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman is typing on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman is typing on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.03160905838012695, 0.10196208953857422], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.03160905838012695, 0.10196208953857422], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003166.pt"}
{"idx": 3168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gyktHss8qmU.mp4", "duration": 59.893166666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time can we see a close-up of the robot's wheels on the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time can we see a close-up of the robot's wheels on the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.8330360651016235, -0.7328577041625977], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.39892977476119995, -0.2987513542175293], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003168.pt"}
{"idx": 3170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o2AsIXSh2xo.mp4", "duration": 176.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman smashes her head against a glass panel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman smashes her head against a glass panel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 50.0], "base_se_norm": [-0.4574993848800659, -0.43489521741867065], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.4914056658744812, -0.46880149841308594], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003170.pt"}
{"idx": 3172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o2AsIXSh2xo.mp4", "duration": 176.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the car flip over in a crash?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the car flip over in a crash?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 79.0], "base_se_norm": [-0.14104068279266357, -0.10713440179824829], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [-0.1523427963256836, -0.11843651533126831], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003172.pt"}
{"idx": 3174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o2AsIXSh2xo.mp4", "duration": 176.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a blindfolded woman and two children are on a boat in turbulent waters.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a blindfolded woman and two children are on a boat in turbulent waters.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 151.0], "base_se_norm": [0.3449493646621704, 0.7066165208816528], "gt_se_raw": [120.0, 122.0], "gt_se_norm": [0.35625147819519043, 0.37885570526123047], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003174.pt"}
{"idx": 3176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xyw5fd5H2j0.mp4", "duration": 58.625233333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are two people playing by rubbing their hands together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two people playing by rubbing their hands together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 29.0], "base_se_norm": [-0.3859299421310425, -0.010664880275726318], "gt_se_raw": [17.0, 28.0], "gt_se_norm": [-0.4200449585914612, -0.04477989673614502], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003176.pt"}
{"idx": 3178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xyw5fd5H2j0.mp4", "duration": 58.625233333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman kisses the other person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman kisses the other person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 50.0], "base_se_norm": [0.26225507259368896, 0.7057501077651978], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [0.535175085067749, 0.6034051179885864], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003178.pt"}
{"idx": 3180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_10tsL37VS4.mp4", "duration": 86.58333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the child kicking both legs up and down rapidly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the child kicking both legs up and down rapidly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.9307026267051697, -0.7690086364746094], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7921078205108643, -0.699711263179779], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003180.pt"}
{"idx": 3182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_10tsL37VS4.mp4", "duration": 86.58333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the child kicks the blue and green spherical toy, making it swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the child kicks the blue and green spherical toy, making it swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.3628488779067993, 0.45524537563323975], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.2608277201652527, -0.21462947130203247], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003182.pt"}
{"idx": 3184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_10tsL37VS4.mp4", "duration": 86.58333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the child turns onto their left side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the child turns onto their left side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.376323401927948, -0.3070260286331177], "gt_se_raw": [74.0, 77.0], "gt_se_norm": [0.7093358039855957, 0.7786332368850708], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003184.pt"}
{"idx": 3186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xct-dgUvlp8.mp4", "duration": 27.133333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog twitch its paws while dreaming?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog twitch its paws while dreaming?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.7051596641540527, -0.2628992795944214], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7788697481155396, -0.5577395558357239], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003186.pt"}
{"idx": 3188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xct-dgUvlp8.mp4", "duration": 27.133333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the dog vigorously rolls its body.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the dog vigorously rolls its body.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [0.1056511402130127, 0.547911524772644], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.4742014408111572, 0.6953316926956177], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003188.pt"}
{"idx": 3190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3YE4sWGC2AM.mp4", "duration": 33.833800000000004, "prompt_text": "Please find the visual event described by the sentence 'When are the man and the boy sitting at the table together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and the boy sitting at the table together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 12.0], "base_se_norm": [-1.0, -0.2906501889228821], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, -0.3497626781463623], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003190.pt"}
{"idx": 3192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3YE4sWGC2AM.mp4", "duration": 33.833800000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the man lift the boy into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man lift the boy into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [0.06402468681335449, 0.3004746437072754], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [0.06402468681335449, 0.3004746437072754], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003192.pt"}
{"idx": 3194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3YE4sWGC2AM.mp4", "duration": 33.833800000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man place the boy back on his feet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man place the boy back on his feet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.7142620086669922, 0.8324869871139526], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [0.7142620086669922, 0.8324869871139526], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003194.pt"}
{"idx": 3196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tUyeaT2ZX1I.mp4", "duration": 169.23573333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a man is working in a call center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a man is working in a call center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.5391044616699219, -0.4918330907821655], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.5509222745895386, -0.5036509037017822], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003196.pt"}
{"idx": 3198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tUyeaT2ZX1I.mp4", "duration": 169.23573333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man assemble electronic components with red and yellow wires?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man assemble electronic components with red and yellow wires?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 68.0], "base_se_norm": [-0.3145655393600464, -0.19638723134994507], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [-0.290929913520813, -0.23184072971343994], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003198.pt"}
{"idx": 3200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ysm7LT3ojDQ.mp4", "duration": 156.23941666666664, "prompt_text": "Please find the visual event described by the sentence 'When are the two children playing with a toy unicorn and drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two children playing with a toy unicorn and drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9615973830223083, -0.897593080997467], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9743982553482056, -0.9487965106964111], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003200.pt"}
{"idx": 3202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ysm7LT3ojDQ.mp4", "duration": 156.23941666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy comes downstairs and joins his dad and sister on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy comes downstairs and joins his dad and sister on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 122.0], "base_se_norm": [0.5233032703399658, 0.5617058277130127], "gt_se_raw": [118.0, 122.0], "gt_se_norm": [0.5105023384094238, 0.5617058277130127], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003202.pt"}
{"idx": 3204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ysm7LT3ojDQ.mp4", "duration": 156.23941666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the youngest girl have purple marker on her face while playing the drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the youngest girl have purple marker on her face while playing the drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 139.0], "base_se_norm": [0.7281171083450317, 0.7793205976486206], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [0.7409180402755737, 0.7665196657180786], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003204.pt"}
{"idx": 3206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zC2G6lf9fCs.mp4", "duration": 60.06, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man put the paper gown over his suit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man put the paper gown over his suit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.6669996976852417, -0.5337995290756226], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.6336996555328369, -0.5337995290756226], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003206.pt"}
{"idx": 3208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zC2G6lf9fCs.mp4", "duration": 60.06, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the patient in the gown is arranging the blood test tubes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the patient in the gown is arranging the blood test tubes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [0.29870128631591797, 0.4319014549255371], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [0.29870128631591797, 0.4319014549255371], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003208.pt"}
{"idx": 3210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7wJRtJvcyto.mp4", "duration": 150.31683333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip introducing the television series Vikings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip introducing the television series Vikings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 8.0], "base_se_norm": [-0.9733895659446716, -0.893558144569397], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9866947531700134, -0.9068633913993835], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003210.pt"}
{"idx": 3212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7wJRtJvcyto.mp4", "duration": 150.31683333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the part where two men are practicing sword fighting in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where two men are practicing sword fighting in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 30.0], "base_se_norm": [-0.7072849273681641, -0.600843071937561], "gt_se_raw": [25.0, 30.0], "gt_se_norm": [-0.6673692464828491, -0.600843071937561], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003212.pt"}
{"idx": 3214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7wJRtJvcyto.mp4", "duration": 150.31683333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the logo featuring a stylized horse and rider above the letters RDA.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the logo featuring a stylized horse and rider above the letters RDA.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.5210117101669312, -0.46779078245162964], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.5210117101669312, -0.4810960292816162], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003214.pt"}
{"idx": 3216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZGA5yi-_QHM.mp4", "duration": 81.68, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cowboy toy get carried away by a red balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cowboy toy get carried away by a red balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.6571987867355347, -0.5592555999755859], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.6571987867355347, -0.5592555999755859], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003216.pt"}
{"idx": 3218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZGA5yi-_QHM.mp4", "duration": 81.68, "prompt_text": "Please find the visual event described by the sentence 'When do the two toys land on the rooftop together at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two toys land on the rooftop together at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.17531836032867432, 0.27326154708862305], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.19980406761169434, 0.27326154708862305], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003218.pt"}
{"idx": 3220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7IhJBWv4Qx8.mp4", "duration": 154.88806666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a close-up of a visual schedule board with pictures and words?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a close-up of a visual schedule board with pictures and words?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 72.0], "base_se_norm": [-0.19942182302474976, -0.07029634714126587], "gt_se_raw": [63.0, 72.0], "gt_se_norm": [-0.18650931119918823, -0.07029634714126587], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003220.pt"}
{"idx": 3222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7IhJBWv4Qx8.mp4", "duration": 154.88806666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman holding a man's arm while looking at a communication device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman holding a man's arm while looking at a communication device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 133.0], "base_se_norm": [0.5624186992645264, 0.7173693180084229], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.5753312110900879, 0.614068865776062], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003222.pt"}
{"idx": 3224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7IhJBWv4Qx8.mp4", "duration": 154.88806666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the woman starts laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the woman starts laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 110.0], "base_se_norm": [0.3816429376602173, 0.4203805923461914], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.3816429376602173, 0.4074680805206299], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003224.pt"}
{"idx": 3226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PmFyn_0udo8.mp4", "duration": 469.84437499999996, "prompt_text": "Please find the visual event described by the sentence 'When do the two men appear outdoors at the end of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 439.0 - 463.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two men appear outdoors at the end of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 439.0 - 463.0 seconds.<|im_end|>\n", "base_se_raw": [439.0, 463.0], "base_se_norm": [0.8687039613723755, 0.9708653688430786], "gt_se_raw": [439.0, 462.0], "gt_se_norm": [0.8687039613723755, 0.9666086435317993], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003226.pt"}
{"idx": 3228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PmFyn_0udo8.mp4", "duration": 469.84437499999996, "prompt_text": "Please find the visual event described by the sentence 'At what point is an animation of a person reading on a beach shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 356.0 - 362.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an animation of a person reading on a beach shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 356.0 - 362.0 seconds.<|im_end|>\n", "base_se_raw": [356.0, 362.0], "base_se_norm": [0.5153954029083252, 0.540935754776001], "gt_se_raw": [356.0, 361.0], "gt_se_norm": [0.5153954029083252, 0.5366790294647217], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003228.pt"}
{"idx": 3230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8pfPl8BkfVY.mp4", "duration": 122.95616666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man with a paper bag over his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man with a paper bag over his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 18.0], "base_se_norm": [-0.902404248714447, -0.7072127461433411], "gt_se_raw": [7.0, 17.0], "gt_se_norm": [-0.8861383199691772, -0.7234786748886108], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003230.pt"}
{"idx": 3232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8pfPl8BkfVY.mp4", "duration": 122.95616666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman shapes the man's beard into a mustache.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman shapes the man's beard into a mustache.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 59.0], "base_se_norm": [-0.18670207262039185, -0.04030841588973999], "gt_se_raw": [50.0, 54.0], "gt_se_norm": [-0.18670207262039185, -0.1216382384300232], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003232.pt"}
{"idx": 3234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8pfPl8BkfVY.mp4", "duration": 122.95616666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man get angry and flip the chair over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get angry and flip the chair over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 81.0], "base_se_norm": [0.2687448263168335, 0.3175426721572876], "gt_se_raw": [78.0, 80.0], "gt_se_norm": [0.2687448263168335, 0.30127668380737305], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003234.pt"}
{"idx": 3236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cYNdUM2gRsg.mp4", "duration": 93.76, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a skater rides into a minefield and explodes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a skater rides into a minefield and explodes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.4240614175796509, -0.36006826162338257], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.4240614175796509, -0.38139933347702026], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003236.pt"}
{"idx": 3238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cYNdUM2gRsg.mp4", "duration": 93.76, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the time machine activates and a dinosaur appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the time machine activates and a dinosaur appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [0.002559661865234375, 0.10921502113342285], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.02389073371887207, 0.10921502113342285], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003238.pt"}
{"idx": 3240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ALXYLq5STlM.mp4", "duration": 43.076366666666665, "prompt_text": "Please find the visual event described by the sentence 'When are the woman and child singing a song together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman and child singing a song together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 32.0], "base_se_norm": [-0.9535708427429199, 0.4857335090637207], "gt_se_raw": [15.0, 38.0], "gt_se_norm": [-0.3035624027252197, 0.7643085718154907], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003240.pt"}
{"idx": 3242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ALXYLq5STlM.mp4", "duration": 43.076366666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the child shouts excitedly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the child shouts excitedly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.06787097454071045, 0.20715844631195068], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.06787097454071045, 0.1607292890548706], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003242.pt"}
{"idx": 3244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ALXYLq5STlM.mp4", "duration": 43.076366666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman leans in to kiss the child's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman leans in to kiss the child's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.8142833113670349, -0.5357083082199097], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8142833113670349, -0.72142493724823], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003244.pt"}
{"idx": 3246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fuj29nJezzc.mp4", "duration": 335.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a coach wearing a gray jacket and BOSE headphones is looking down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a coach wearing a gray jacket and BOSE headphones is looking down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.9224266409873962, -0.8985579013824463], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.9224266409873962, -0.8985579013824463], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003246.pt"}
{"idx": 3248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fuj29nJezzc.mp4", "duration": 335.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the coach in the white shirt use a folder to cover his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the coach in the white shirt use a folder to cover his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 121.0], "base_se_norm": [-0.3018398880958557, -0.27797114849090576], "gt_se_raw": [117.0, 121.0], "gt_se_norm": [-0.3018398880958557, -0.27797114849090576], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003248.pt"}
{"idx": 3250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbtsKmUMVhU.mp4", "duration": 19.586233333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the person first places the tool between their feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the person first places the tool between their feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.5915498733520508], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.5915498733520508], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003250.pt"}
{"idx": 3252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbtsKmUMVhU.mp4", "duration": 19.586233333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the person on the right touch their face with their hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person on the right touch their face with their hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [0.2253504991531372, 0.838025689125061], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [0.2253504991531372, 0.5316880941390991], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003252.pt"}
{"idx": 3254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbtsKmUMVhU.mp4", "duration": 19.586233333333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the person begins the attempt to pick up the white circular object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the person begins the attempt to pick up the white circular object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.1830996870994568, 0.2253504991531372], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.1830996870994568, 0.1232379674911499], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003254.pt"}
{"idx": 3256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5xJ6h6OTveU.mp4", "duration": 82.48248248248248, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the winning team starts celebrating together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the winning team starts celebrating together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 28.0], "base_se_norm": [-0.5635436773300171, -0.3210679888725281], "gt_se_raw": [15.0, 23.0], "gt_se_norm": [-0.6362863779067993, -0.4423058032989502], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003256.pt"}
{"idx": 3258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5xJ6h6OTveU.mp4", "duration": 82.48248248248248, "prompt_text": "Please find the visual event described by the sentence 'Locate the slow-motion replay of the final shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the slow-motion replay of the final shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 82.0], "base_se_norm": [0.6730824708938599, 0.9883009195327759], "gt_se_raw": [69.0, 77.0], "gt_se_norm": [0.6730824708938599, 0.867063045501709], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003258.pt"}
{"idx": 3260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S8hUzi6v43c.mp4", "duration": 189.23070833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the 20-pound batches poured onto the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the 20-pound batches poured onto the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.4504063129425049, -0.4081299304962158], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [-0.4504063129425049, -0.4186990261077881], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003260.pt"}
{"idx": 3262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S8hUzi6v43c.mp4", "duration": 189.23070833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the candy being stretched on a hook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the candy being stretched on a hook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 104.0], "base_se_norm": [0.05691087245941162, 0.09918737411499023], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [0.06747996807098389, 0.1097564697265625], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003262.pt"}
{"idx": 3264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S8hUzi6v43c.mp4", "duration": 189.23070833333333, "prompt_text": "Please find the visual event described by the sentence 'What is the process of heating the candy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the process of heating the candy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 47.0], "base_se_norm": [-0.6829267740249634, -0.5032519102096558], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.6829267740249634, -0.651219367980957], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003264.pt"}
{"idx": 3266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W9fRvESVClE.mp4", "duration": 106.29166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the dog pick up the photo that fell from the zombie's pocket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog pick up the photo that fell from the zombie's pocket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 25.0], "base_se_norm": [-0.6048608422279358, -0.529596209526062], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [-0.5672285556793213, -0.5107800960540771], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003266.pt"}
{"idx": 3268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W9fRvESVClE.mp4", "duration": 106.29166666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the zombie throw its severed arm to play fetch with the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the zombie throw its severed arm to play fetch with the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [0.09133672714233398, 0.16660141944885254], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [0.1101529598236084, 0.14778518676757812], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003268.pt"}
{"idx": 3270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_F6h0yH7EyU.mp4", "duration": 145.25, "prompt_text": "Please find the visual event described by the sentence 'When is the Grinch posing in front of the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the Grinch posing in front of the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9586918950080872], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9586918950080872], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003270.pt"}
{"idx": 3272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_F6h0yH7EyU.mp4", "duration": 145.25, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the dog makes a sad face in the coffee foam.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the dog makes a sad face in the coffee foam.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 66.0], "base_se_norm": [-0.11876076459884644, -0.09122204780578613], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [-0.1049913763999939, -0.0774526596069336], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003272.pt"}
{"idx": 3274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_F6h0yH7EyU.mp4", "duration": 145.25, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the Grinch breaks a jar in the supermarket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the Grinch breaks a jar in the supermarket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 107.0], "base_se_norm": [0.4457831382751465, 0.47332191467285156], "gt_se_raw": [102.0, 106.0], "gt_se_norm": [0.4044750928878784, 0.459552526473999], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003274.pt"}
{"idx": 3276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICVXf8Vznec.mp4", "duration": 169.40256666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man is working on the bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man is working on the bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 34.0], "base_se_norm": [-0.7638760805130005, -0.598589301109314], "gt_se_raw": [20.0, 37.0], "gt_se_norm": [-0.7638760805130005, -0.5631707310676575], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003276.pt"}
{"idx": 3278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICVXf8Vznec.mp4", "duration": 169.40256666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point do the parents comfort their crying daughter on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the parents comfort their crying daughter on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 118.0], "base_se_norm": [0.08617007732391357, 0.3931311368942261], "gt_se_raw": [92.0, 118.0], "gt_se_norm": [0.08617007732391357, 0.3931311368942261], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003278.pt"}
{"idx": 3280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/baNaqhoBjUE.mp4", "duration": 198.96543333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the boy put the cards on the board to form a sentence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy put the cards on the board to form a sentence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 62.0], "base_se_norm": [-0.6582320928573608, -0.37677615880966187], "gt_se_raw": [5.0, 14.0], "gt_se_norm": [-0.9497399926185608, -0.8592720627784729], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003280.pt"}
{"idx": 3282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/baNaqhoBjUE.mp4", "duration": 198.96543333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy touch the symbol for bread?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy touch the symbol for bread?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 41.0], "base_se_norm": [-0.6582320928573608, -0.5878680944442749], "gt_se_raw": [46.0, 55.0], "gt_se_norm": [-0.5376081466674805, -0.4471401572227478], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003282.pt"}
{"idx": 3284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/baNaqhoBjUE.mp4", "duration": 198.96543333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a striped sweater reading the book with the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a striped sweater reading the book with the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 198.0], "base_se_norm": [0.19618761539459229, 0.99029541015625], "gt_se_raw": [118.0, 128.0], "gt_se_norm": [0.1861356496810913, 0.2866556644439697], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003284.pt"}
{"idx": 3286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vbuq7w3ZDUQ.mp4", "duration": 416.99991666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the little boy set foot on the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little boy set foot on the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 163.0], "base_se_norm": [-0.23740988969802856, -0.21822524070739746], "gt_se_raw": [155.0, 166.0], "gt_se_norm": [-0.25659453868865967, -0.20383673906326294], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003286.pt"}
{"idx": 3288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vbuq7w3ZDUQ.mp4", "duration": 416.99991666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the boy hit the big star with a hammer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 318.0 - 321.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy hit the big star with a hammer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 318.0 - 321.0 seconds.<|im_end|>\n", "base_se_raw": [318.0, 321.0], "base_se_norm": [0.5251802206039429, 0.5395686626434326], "gt_se_raw": [318.0, 321.0], "gt_se_norm": [0.5251802206039429, 0.5395686626434326], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003288.pt"}
{"idx": 3290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xl6yy6a3emw.mp4", "duration": 170.5704, "prompt_text": "Please find the visual event described by the sentence 'When does the man enter the room and close the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man enter the room and close the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9648239016532898], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.9413731694221497], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003290.pt"}
{"idx": 3292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xl6yy6a3emw.mp4", "duration": 170.5704, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man shouts at the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man shouts at the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 87.0], "base_se_norm": [-0.09714698791503906, 0.02010667324066162], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [-0.09714698791503906, -0.07369625568389893], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003292.pt"}
{"idx": 3294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xl6yy6a3emw.mp4", "duration": 170.5704, "prompt_text": "Please find the visual event described by the sentence 'When do the man and woman start kissing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the man and woman start kissing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 123.0], "base_se_norm": [0.3835928440093994, 0.44221973419189453], "gt_se_raw": [118.0, 123.0], "gt_se_norm": [0.3835928440093994, 0.44221973419189453], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003294.pt"}
{"idx": 3296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p_Rrovk5nsk.mp4", "duration": 64.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the person turns on the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the person turns on the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.6594427227973938, -0.5665634870529175], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.6284829378128052, -0.5356037020683289], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003296.pt"}
{"idx": 3298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p_Rrovk5nsk.mp4", "duration": 64.6, "prompt_text": "Please find the visual event described by the sentence 'At what point are eggshells seen in a plant pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are eggshells seen in a plant pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 51.0], "base_se_norm": [0.3003096580505371, 0.5789474248886108], "gt_se_raw": [42.0, 51.0], "gt_se_norm": [0.3003096580505371, 0.5789474248886108], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003298.pt"}
{"idx": 3300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzZuNeRdFIc.mp4", "duration": 134.759625, "prompt_text": "Please find the visual event described by the sentence 'When can we see the animated intro of 'BADLANDS CHUGS'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the animated intro of 'BADLANDS CHUGS'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.8515875935554504, -0.777381420135498], "gt_se_raw": [10.0, 16.0], "gt_se_norm": [-0.8515875935554504, -0.7625401616096497], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003300.pt"}
{"idx": 3302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzZuNeRdFIc.mp4", "duration": 134.759625, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man starts drinking the milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man starts drinking the milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.1540493369102478, -0.07984310388565063], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.13920807838439941, -0.09468436241149902], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003302.pt"}
{"idx": 3304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzZuNeRdFIc.mp4", "duration": 134.759625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man celebrate by yelling after finishing the milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man celebrate by yelling after finishing the milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 107.0], "base_se_norm": [0.5434889793395996, 0.5880126953125], "gt_se_raw": [105.0, 107.0], "gt_se_norm": [0.5583301782608032, 0.5880126953125], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003304.pt"}
{"idx": 3306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vsmBpaVqhmc.mp4", "duration": 181.01416666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the man seen running through the museum lobby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man seen running through the museum lobby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.7458761930465698, -0.7127296924591064], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.7348273992538452, -0.7127296924591064], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003306.pt"}
{"idx": 3308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vsmBpaVqhmc.mp4", "duration": 181.01416666666668, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man touch his chin thoughtfully?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man touch his chin thoughtfully?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 105.0], "base_se_norm": [0.10488593578338623, 0.16013026237487793], "gt_se_raw": [100.0, 103.0], "gt_se_norm": [0.10488593578338623, 0.13803255558013916], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003308.pt"}
{"idx": 3310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qCNodlSc6Hw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man pouring a white substance into a bowl on a scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man pouring a white substance into a bowl on a scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8032602071762085], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003310.pt"}
{"idx": 3312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qCNodlSc6Hw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman looking into a microscope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman looking into a microscope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003312.pt"}
{"idx": 3314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qCNodlSc6Hw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When are two people in lab coats working together with a precision scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two people in lab coats working together with a precision scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003314.pt"}
{"idx": 3316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9qlF_9PNhJs.mp4", "duration": 138.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man get stuck on the tree while hugging the tree trunk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man get stuck on the tree while hugging the tree trunk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 52.0], "base_se_norm": [-0.4929006099700928, -0.24659514427185059], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.5363662242889404, -0.44943493604660034], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003316.pt"}
{"idx": 3318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9qlF_9PNhJs.mp4", "duration": 138.04, "prompt_text": "Please find the visual event described by the sentence 'When does the cat begin to descend the tree on its own?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat begin to descend the tree on its own?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 86.0], "base_se_norm": [0.11561870574951172, 0.24601566791534424], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [0.21703863143920898, 0.26050424575805664], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003318.pt"}
{"idx": 3320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLdbbZZRvtM.mp4", "duration": 37.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man introduce himself and explain the origin of his head injuries?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man introduce himself and explain the origin of his head injuries?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.2998204827308655], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.7307001948356628], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003320.pt"}
{"idx": 3322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLdbbZZRvtM.mp4", "duration": 37.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when he discusses planning for work he might miss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when he discusses planning for work he might miss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.0843806266784668, 0.23877906799316406], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.07719922065734863, 0.18491911888122559], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003322.pt"}
{"idx": 3324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLdbbZZRvtM.mp4", "duration": 37.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the specific time he concludes by restating his primary advice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the specific time he concludes by restating his primary advice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 36.0], "base_se_norm": [0.615798830986023, 0.9389586448669434], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [0.7773786783218384, 0.9389586448669434], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003324.pt"}
{"idx": 3326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JerVrbLldXw.mp4", "duration": 150.275125, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is in a machine being transformed into a super-soldier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is in a machine being transformed into a super-soldier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 93.0], "base_se_norm": [-0.2680093050003052, 0.23772978782653809], "gt_se_raw": [65.0, 92.0], "gt_se_norm": [-0.13492006063461304, 0.22442078590393066], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003326.pt"}
{"idx": 3328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JerVrbLldXw.mp4", "duration": 150.275125, "prompt_text": "Please find the visual event described by the sentence 'When does the villain pull off his face to reveal a red skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the villain pull off his face to reveal a red skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 115.0], "base_se_norm": [0.5039081573486328, 0.5305260419845581], "gt_se_raw": [113.0, 114.0], "gt_se_norm": [0.5039081573486328, 0.5172171592712402], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003328.pt"}
{"idx": 3330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epKCqDN9fBo.mp4", "duration": 190.39854166666666, "prompt_text": "Please find the visual event described by the sentence 'When are the two women walking and conversing on the running track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women walking and conversing on the running track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 127.0], "base_se_norm": [0.2184966802597046, 0.33404386043548584], "gt_se_raw": [117.0, 122.0], "gt_se_norm": [0.22900104522705078, 0.28152239322662354], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003330.pt"}
{"idx": 3332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epKCqDN9fBo.mp4", "duration": 190.39854166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dog shown lying on the floor in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dog shown lying on the floor in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 69.0], "base_se_norm": [-0.36974310874938965, -0.27520453929901123], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [-0.36974310874938965, -0.31722164154052734], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003332.pt"}
{"idx": 3334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epKCqDN9fBo.mp4", "duration": 190.39854166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a group of runners starts the race.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a group of runners starts the race.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 154.0], "base_se_norm": [0.5651379823684692, 0.6176594495773315], "gt_se_raw": [149.0, 153.0], "gt_se_norm": [0.5651379823684692, 0.6071552038192749], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003334.pt"}
{"idx": 3336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vxd9W0RVgFY.mp4", "duration": 91.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man enters a bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man enters a bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.3628707528114319, -0.27499085664749146], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.4068107008934021, -0.3628707528114319], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003336.pt"}
{"idx": 3338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vxd9W0RVgFY.mp4", "duration": 91.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in sunglasses hug the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in sunglasses hug the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 55.0], "base_se_norm": [0.16440868377685547, 0.20834863185882568], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [0.16440868377685547, 0.20834863185882568], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003338.pt"}
{"idx": 3340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-wc7bB2G94.mp4", "duration": 118.15136666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman is seen sitting on a yellow sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman is seen sitting on a yellow sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 57.0], "base_se_norm": [-0.7799432873725891, -0.03513598442077637], "gt_se_raw": [14.0, 56.0], "gt_se_norm": [-0.763015866279602, -0.0520634651184082], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003340.pt"}
{"idx": 3342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-wc7bB2G94.mp4", "duration": 118.15136666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where the woman in the blue jacket is performing a squatting motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where the woman in the blue jacket is performing a squatting motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 43.0], "base_se_norm": [-0.4244670867919922, -0.2721201181411743], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.3059750199317932, -0.25519269704818726], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003342.pt"}
{"idx": 3344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-wc7bB2G94.mp4", "duration": 118.15136666666668, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man in the black t-shirt is speaking while looking towards the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man in the black t-shirt is speaking while looking towards the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 117.0], "base_se_norm": [0.5234694480895996, 0.9805103540420532], "gt_se_raw": [89.0, 117.0], "gt_se_norm": [0.5065419673919678, 0.9805103540420532], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003344.pt"}
{"idx": 3346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2imuIooWjPE.mp4", "duration": 54.72, "prompt_text": "Please find the visual event described by the sentence 'At what time is there a close-up of the yellow sign with pictograms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is there a close-up of the yellow sign with pictograms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.5248538255691528, -0.34210526943206787], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.4152047038078308, -0.34210526943206787], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003346.pt"}
{"idx": 3348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2imuIooWjPE.mp4", "duration": 54.72, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a man bends over to pick up dog droppings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a man bends over to pick up dog droppings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [0.4254385232925415, 0.5350877046585083], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [0.4254385232925415, 0.4985380172729492], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003348.pt"}
{"idx": 3350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymvdlZjewL4.mp4", "duration": 123.04166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the doctor walks into the room and greets the others.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the doctor walks into the room and greets the others.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.869962751865387, -0.8049441576004028], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.869962751865387, -0.8374534249305725], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003350.pt"}
{"idx": 3352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymvdlZjewL4.mp4", "duration": 123.04166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman explain that she feels exhausted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman explain that she feels exhausted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 63.0], "base_se_norm": [-0.21977651119232178, 0.02404332160949707], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.20352184772491455, -0.15475785732269287], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003352.pt"}
{"idx": 3354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymvdlZjewL4.mp4", "duration": 123.04166666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in green interrupt, saying 'hang on' and asking the doctor to look at him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in green interrupt, saying 'hang on' and asking the doctor to look at him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.4473416805267334, -0.3498137593269348], "gt_se_raw": [78.0, 82.0], "gt_se_norm": [0.26786327362060547, 0.3328818082809448], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003354.pt"}
{"idx": 3356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q99rhZlYlAo.mp4", "duration": 110.44, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat transform into a bean shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat transform into a bean shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 36.0], "base_se_norm": [-0.5291560888290405, -0.34806233644485474], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.5110467672348022, -0.34806233644485474], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003356.pt"}
{"idx": 3358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q99rhZlYlAo.mp4", "duration": 110.44, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the cat is doing the 'Flying Pigeon' pose and drinking spilled milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the cat is doing the 'Flying Pigeon' pose and drinking spilled milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 59.0], "base_se_norm": [-0.05831223726272583, 0.06845343112945557], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [-0.05831223726272583, 0.050344109535217285], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003358.pt"}
{"idx": 3360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_JmA2ClUvUY.mp4", "duration": 127.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a baby touching another baby's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a baby touching another baby's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9843668341636658, -0.8436685800552368], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.609171450138092, -0.5466388463973999], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003360.pt"}
{"idx": 3362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_JmA2ClUvUY.mp4", "duration": 127.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the babies are trying to high five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the babies are trying to high five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 89.0], "base_se_norm": [0.2975507974624634, 0.3913496732711792], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [0.3444502353668213, 0.3913496732711792], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003362.pt"}
{"idx": 3364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_JmA2ClUvUY.mp4", "duration": 127.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when a baby kicks up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a baby kicks up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.37467432022094727, -0.32777488231658936], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.7029702663421631, -0.6560708284378052], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003364.pt"}
{"idx": 3366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mm3rTwAxH20.mp4", "duration": 70.90416666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the man shouting by the campfire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man shouting by the campfire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 48.0], "base_se_norm": [-0.15378737449645996, 0.35394012928009033], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.15378737449645996, -0.09737324714660645], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003366.pt"}
{"idx": 3368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mm3rTwAxH20.mp4", "duration": 70.90416666666667, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the woman in bed say 'Not at this time'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the woman in bed say 'Not at this time'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [0.43856143951416016, 0.5795968770980835], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [0.43856143951416016, 0.49497556686401367], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003368.pt"}
{"idx": 3370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O8ct0a3hStg.mp4", "duration": 20.2, "prompt_text": "Please find the visual event described by the sentence 'When is the person getting food from the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person getting food from the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.7029702663421631], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.7029702663421631], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003370.pt"}
{"idx": 3372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O8ct0a3hStg.mp4", "duration": 20.2, "prompt_text": "Please find the visual event described by the sentence 'At what point is the food being poured into the green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the food being poured into the green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [0.2871286869049072, 0.48514842987060547], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.2871286869049072, 0.48514842987060547], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003372.pt"}
{"idx": 3374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O8ct0a3hStg.mp4", "duration": 20.2, "prompt_text": "Please find the visual event described by the sentence 'When is the cat walking from one room to another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cat walking from one room to another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.7029702663421631, -0.10891091823577881], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.603960394859314, -0.2079208493232727], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003374.pt"}
{"idx": 3376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6MW7bkk3MZ0.mp4", "duration": 141.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the yellow larva licks the frozen red larva.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the yellow larva licks the frozen red larva.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [0.006497383117675781, 0.0632014274597168], "gt_se_raw": [72.0, 75.0], "gt_se_norm": [0.020673394203186035, 0.0632014274597168], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003376.pt"}
{"idx": 3378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6MW7bkk3MZ0.mp4", "duration": 141.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the red larva uses a stick to drill a hole in the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the red larva uses a stick to drill a hole in the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 65.0], "base_se_norm": [-0.4471352696418762, -0.07855874300003052], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [-0.13526278734207153, -0.10691076517105103], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003378.pt"}
{"idx": 3380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yhofIxEfld0.mp4", "duration": 13.079733333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the pink liquid poured onto the black fabric?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the pink liquid poured onto the black fabric?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.8470916748046875, -0.38836669921875], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.8470916748046875, -0.5412750244140625], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003380.pt"}
{"idx": 3382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yhofIxEfld0.mp4", "duration": 13.079733333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the hands are rubbing the black fabric.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the hands are rubbing the black fabric.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [0.0703582763671875, 0.529083251953125], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.2232666015625, 0.3761749267578125], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003382.pt"}
{"idx": 3384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yhofIxEfld0.mp4", "duration": 13.079733333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the transparent tube is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the transparent tube is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.8470916748046875, -0.38836669921875], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.5412750244140625], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003384.pt"}
{"idx": 3386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qMzt3yQFT-Q.mp4", "duration": 397.23016666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man with long curly hair is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 184.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man with long curly hair is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 184.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 184.0], "base_se_norm": [-0.1289684772491455, -0.07358497381210327], "gt_se_raw": [174.0, 176.0], "gt_se_norm": [-0.12393361330032349, -0.11386388540267944], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003386.pt"}
{"idx": 3388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qMzt3yQFT-Q.mp4", "duration": 397.23016666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the three men are huddling together and whispering.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 346.0 - 354.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the three men are huddling together and whispering.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 346.0 - 354.0 seconds.<|im_end|>\n", "base_se_raw": [346.0, 354.0], "base_se_norm": [0.742063045501709, 0.7823419570922852], "gt_se_raw": [153.0, 157.0], "gt_se_norm": [-0.22966575622558594, -0.20952630043029785], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003388.pt"}
{"idx": 3390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZUJJdeByJa0.mp4", "duration": 182.28, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the allspice in a small bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the allspice in a small bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8573622703552246, -0.824445903301239], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.8683344125747681, -0.824445903301239], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003390.pt"}
{"idx": 3392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZUJJdeByJa0.mp4", "duration": 182.28, "prompt_text": "Please find the visual event described by the sentence 'At what point is the marinade being blended in the food processor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the marinade being blended in the food processor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 88.0], "base_se_norm": [-0.15514594316482544, -0.03445249795913696], "gt_se_raw": [78.0, 87.0], "gt_se_norm": [-0.14417380094528198, -0.04542464017868042], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003392.pt"}
{"idx": 3394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZUJJdeByJa0.mp4", "duration": 182.28, "prompt_text": "Please find the visual event described by the sentence 'When is the marinated chicken shown cooking on the barbecue grill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the marinated chicken shown cooking on the barbecue grill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 168.0], "base_se_norm": [0.5141540765762329, 0.843317985534668], "gt_se_raw": [149.0, 168.0], "gt_se_norm": [0.6348475217819214, 0.843317985534668], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003394.pt"}
{"idx": 3396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wJ-mdxp_TY0.mp4", "duration": 62.77104166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman has her arms crossed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman has her arms crossed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.6495199203491211, -0.5220726132392883], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.6495199203491211, -0.5539344549179077], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003396.pt"}
{"idx": 3398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wJ-mdxp_TY0.mp4", "duration": 62.77104166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do the flowers on the table transform into a piece of wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the flowers on the table transform into a piece of wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [0.08330202102661133, 0.1470257043838501], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [0.08330202102661133, 0.1470257043838501], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003398.pt"}
{"idx": 3400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_dGXDRZEkc.mp4", "duration": 25.025000000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the little girl successfully press the button on the toy, causing it to light up and make a sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little girl successfully press the button on the toy, causing it to light up and make a sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.6803196668624878], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.3606393337249756, -0.20079916715621948], "start_token_span": [77, 80], "end_token_span": [82, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003400.pt"}
{"idx": 3402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_dGXDRZEkc.mp4", "duration": 25.025000000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point can we hear the sound of crinkling paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we hear the sound of crinkling paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.20079916715621948, 0.11888110637664795], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [0.5184814929962158, 0.6783217191696167], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003402.pt"}
{"idx": 3404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_dGXDRZEkc.mp4", "duration": 25.025000000000002, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the child is lying on the floor with a pacifier in her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the child is lying on the floor with a pacifier in her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, 0.9180819988250732], "gt_se_raw": [0.0, 25.0], "gt_se_norm": [-1.0, 0.9980020523071289], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003404.pt"}
{"idx": 3406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RtZZZmmdgU.mp4", "duration": 146.07926666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the audience clapping and cheering loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the audience clapping and cheering loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 40.0], "base_se_norm": [-0.47973453998565674, -0.4523521065711975], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [-0.47973453998565674, -0.4523521065711975], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003406.pt"}
{"idx": 3408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RtZZZmmdgU.mp4", "duration": 146.07926666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man is hit by the big splash of water from all the balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man is hit by the big splash of water from all the balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 127.0], "base_se_norm": [0.6840171813964844, 0.7387819290161133], "gt_se_raw": [123.0, 126.0], "gt_se_norm": [0.6840171813964844, 0.725090742111206], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003408.pt"}
{"idx": 3410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bwzLiQZDw2I.mp4", "duration": 141.18270833333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see two young sisters sitting on a bed and listening to a story?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two young sisters sitting on a bed and listening to a story?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.9008377194404602, -0.8583396077156067], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.915003776550293, -0.8725056648254395], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003410.pt"}
{"idx": 3412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bwzLiQZDw2I.mp4", "duration": 141.18270833333332, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the group entering the misty enchanted forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the group entering the misty enchanted forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [0.09078514575958252, 0.16161537170410156], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [-0.008377134799957275, 0.019954919815063477], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003412.pt"}
{"idx": 3414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bwzLiQZDw2I.mp4", "duration": 141.18270833333332, "prompt_text": "Please find the visual event described by the sentence 'When does Elsa run across the dark sea, creating an icy path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Elsa run across the dark sea, creating an icy path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.5041885375976562, -0.4191923141479492], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.5157663822174072, 0.5440984964370728], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003414.pt"}
{"idx": 3416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IASj8rqh24c.mp4", "duration": 140.10471204188482, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker mentions the two biggest things she struggles with regarding sleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker mentions the two biggest things she struggles with regarding sleep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 22.0], "base_se_norm": [-0.8572496175765991, -0.6859492063522339], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.543198823928833, -0.4718235731124878], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003416.pt"}
{"idx": 3418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IASj8rqh24c.mp4", "duration": 140.10471204188482, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the woman shows the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the woman shows the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 60.0], "base_se_norm": [-0.2434229850769043, -0.14349770545959473], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.17204779386520386, -0.14349770545959473], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003418.pt"}
{"idx": 3420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/faYA_upR_a0.mp4", "duration": 127.76, "prompt_text": "Please find the visual event described by the sentence 'When does the man place the cake box on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man place the cake box on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.874765157699585, -0.796493411064148], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.843456506729126, -0.8121477961540222], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003420.pt"}
{"idx": 3422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/faYA_upR_a0.mp4", "duration": 127.76, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man tastes the cake frosting and then runs off in disgust.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man tastes the cake frosting and then runs off in disgust.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 56.0], "base_se_norm": [-0.45209771394729614, -0.12335628271102905], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [-0.17031937837600708, -0.12335628271102905], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003422.pt"}
{"idx": 3424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/faYA_upR_a0.mp4", "duration": 127.76, "prompt_text": "Please find the visual event described by the sentence 'When are both the man and the cat wearing party hats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are both the man and the cat wearing party hats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 96.0], "base_se_norm": [0.29931116104125977, 0.5028177499771118], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [0.31496548652648926, 0.3619285821914673], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003424.pt"}
{"idx": 3426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EAEu4brBayI.mp4", "duration": 47.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the period when the student is viewing a black-and-white illustration on his device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period when the student is viewing a black-and-white illustration on his device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 36.0], "base_se_norm": [-0.20112121105194092, 0.5136650800704956], "gt_se_raw": [20.0, 38.0], "gt_se_norm": [-0.15907496213912964, 0.5977575778961182], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003426.pt"}
{"idx": 3428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EAEu4brBayI.mp4", "duration": 47.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the boy places his right hand on top of the monitor and holds it there.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the boy places his right hand on top of the monitor and holds it there.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [0.5557112693786621, 0.9341275691986084], "gt_se_raw": [37.0, 47.0], "gt_se_norm": [0.5557112693786621, 0.9761738777160645], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003428.pt"}
{"idx": 3430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/813_zdNJYa4.mp4", "duration": 192.42556666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the ingredients for the English muffin pizzas are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the ingredients for the English muffin pizzas are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 39.0], "base_se_norm": [-1.0, -0.5946484804153442], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, -0.7921274304389954], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003430.pt"}
{"idx": 3432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/813_zdNJYa4.mp4", "duration": 192.42556666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the person applies pizza sauce to the muffins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the person applies pizza sauce to the muffins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 93.0], "base_se_norm": [-0.23087143898010254, -0.033392488956451416], "gt_se_raw": [74.0, 93.0], "gt_se_norm": [-0.23087143898010254, -0.033392488956451416], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003432.pt"}
{"idx": 3434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/813_zdNJYa4.mp4", "duration": 192.42556666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the final cooked pizzas on a plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the final cooked pizzas on a plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 185.0], "base_se_norm": [0.6733742952346802, 0.9228214025497437], "gt_se_raw": [161.0, 184.0], "gt_se_norm": [0.6733742952346802, 0.9124277830123901], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003434.pt"}
{"idx": 3436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XHY2eXoCeSY.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where two men are sitting at desks in front of a wall of equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where two men are sitting at desks in front of a wall of equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003436.pt"}
{"idx": 3438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XHY2eXoCeSY.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman with glasses is using a screwdriver.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman with glasses is using a screwdriver.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003438.pt"}
{"idx": 3440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d6g1c18Cy-8.mp4", "duration": 211.97843333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two individuals watching a video on the television?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two individuals watching a video on the television?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 31.0], "base_se_norm": [-0.9905650615692139, -0.7075173854827881], "gt_se_raw": [2.0, 30.0], "gt_se_norm": [-0.9811301827430725, -0.7169523239135742], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003440.pt"}
{"idx": 3442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d6g1c18Cy-8.mp4", "duration": 211.97843333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the man combs his hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the man combs his hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 75.0], "base_se_norm": [-0.36786025762557983, -0.2923808693885803], "gt_se_raw": [68.0, 75.0], "gt_se_norm": [-0.3584253191947937, -0.2923808693885803], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003442.pt"}
{"idx": 3444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d6g1c18Cy-8.mp4", "duration": 211.97843333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the person wearing glasses takes objects out of the black container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the person wearing glasses takes objects out of the black container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 155.0], "base_se_norm": [0.15106046199798584, 0.46241283416748047], "gt_se_raw": [125.0, 155.0], "gt_se_norm": [0.17936527729034424, 0.46241283416748047], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003444.pt"}
{"idx": 3446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7AlNZmmVU7U.mp4", "duration": 63.797066666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman explains the first part of her mission, which involves filming videos.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman explains the first part of her mission, which involves filming videos.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.27896368503570557], "gt_se_raw": [13.0, 30.0], "gt_se_norm": [-0.5924577713012695, -0.059517860412597656], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003446.pt"}
{"idx": 3448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7AlNZmmVU7U.mp4", "duration": 63.797066666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does she list the specific software she plans to install on the computers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she list the specific software she plans to install on the computers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 38.0], "base_se_norm": [-0.247614324092865, 0.1912773847579956], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [0.15992796421051025, 0.2539762258529663], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003448.pt"}
{"idx": 3450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GHDz-XDD8OU.mp4", "duration": 30.932000000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the intruder enter the house through the sliding door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the intruder enter the house through the sliding door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.8706840872764587, -0.6120522022247314], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.8706840872764587, -0.6120522022247314], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003450.pt"}
{"idx": 3452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GHDz-XDD8OU.mp4", "duration": 30.932000000000002, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the intruder take a chip from the sleeping man's bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the intruder take a chip from the sleeping man's bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [0.22850120067596436, 0.4871331453323364], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.29315924644470215, 0.4871331453323364], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003452.pt"}
{"idx": 3454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GHDz-XDD8OU.mp4", "duration": 30.932000000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the intruder is looking through the shelves and cabinets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the intruder is looking through the shelves and cabinets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.5473942756652832, -0.15944647789001465], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.41807836294174194, -0.15944647789001465], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003454.pt"}
{"idx": 3456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TsujuQbDpI0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the statistic '79.1%' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the statistic '79.1%' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9366739988327026], "start_token_span": [78, 81], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003456.pt"}
{"idx": 3458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TsujuQbDpI0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker states that the employment rate for Americans without disabilities is much higher.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker states that the employment rate for Americans without disabilities is much higher.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9031662940979004, -0.3221641182899475], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9031662940979004, -0.4189978241920471], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003458.pt"}
{"idx": 3460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D01ihkEgdho.mp4", "duration": 128.89556222889556, "prompt_text": "Please find the visual event described by the sentence 'When can we see three people collaborating on a task in a bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three people collaborating on a task in a bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.8913849592208862, -0.7827698588371277], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.8913849592208862, -0.7827698588371277], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003460.pt"}
{"idx": 3462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D01ihkEgdho.mp4", "duration": 128.89556222889556, "prompt_text": "Please find the visual event described by the sentence 'At what point do two women in orange vests converse outside the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two women in orange vests converse outside the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.5189904570579529, -0.4724411368370056], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.5189904570579529, -0.4724411368370056], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003462.pt"}
{"idx": 3464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D01ihkEgdho.mp4", "duration": 128.89556222889556, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two women are talking and laughing in a living room setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two women are talking and laughing in a living room setting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 95.0], "base_se_norm": [0.4119957685470581, 0.4740614891052246], "gt_se_raw": [91.0, 94.0], "gt_se_norm": [0.4119957685470581, 0.45854508876800537], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003464.pt"}
{"idx": 3466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtY40NIWg0o.mp4", "duration": 130.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the speaker mention she used a white cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the speaker mention she used a white cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.507944643497467, -0.4464377164840698], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.4618144631385803, -0.4310609698295593], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003466.pt"}
{"idx": 3468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtY40NIWg0o.mp4", "duration": 130.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the woman describes working in Washington D.C. and taking the metro.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the woman describes working in Washington D.C. and taking the metro.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 50.0], "base_se_norm": [-0.507944643497467, -0.23116350173950195], "gt_se_raw": [58.0, 66.0], "gt_se_norm": [-0.10814964771270752, 0.014864206314086914], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003468.pt"}
{"idx": 3470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fmnUwvZAMjQ.mp4", "duration": 181.0, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the text 'Strategy #1'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the text 'Strategy #1'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 27.0], "base_se_norm": [-0.7900552749633789, -0.7016574144363403], "gt_se_raw": [19.0, 27.0], "gt_se_norm": [-0.7900552749633789, -0.7016574144363403], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003470.pt"}
{"idx": 3472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fmnUwvZAMjQ.mp4", "duration": 181.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the woman has a thought bubble above her head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the woman has a thought bubble above her head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 76.0], "base_se_norm": [-0.23756903409957886, -0.16022098064422607], "gt_se_raw": [68.0, 76.0], "gt_se_norm": [-0.24861878156661987, -0.16022098064422607], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003472.pt"}
{"idx": 3474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fmnUwvZAMjQ.mp4", "duration": 181.0, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the 'Final Thoughts' section?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the 'Final Thoughts' section?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 167.0], "base_se_norm": [0.6685082912445068, 0.8453038930892944], "gt_se_raw": [152.0, 167.0], "gt_se_norm": [0.6795580387115479, 0.8453038930892944], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003474.pt"}
{"idx": 3476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xNrR4JMACMg.mp4", "duration": 149.9498, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the presenter is filming in a selfie-style and making a transition in her speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the presenter is filming in a selfie-style and making a transition in her speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 26.0], "base_se_norm": [-1.0, -0.6532173156738281], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.7065684795379639, -0.6398794651031494], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003476.pt"}
{"idx": 3478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xNrR4JMACMg.mp4", "duration": 149.9498, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a list of movies, including 'Stranger Things' and 'Stronger', is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a list of movies, including 'Stranger Things' and 'Stronger', is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 68.0], "base_se_norm": [-0.21306997537612915, -0.09302979707717896], "gt_se_raw": [60.0, 69.0], "gt_se_norm": [-0.19973218441009521, -0.07969200611114502], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003478.pt"}
{"idx": 3480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTxn2BuqyzU.mp4", "duration": 176.04166666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy eat the watermelon slice, including the seed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy eat the watermelon slice, including the seed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6932544708251953, -0.6591715812683105], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.6932544708251953, -0.6591715812683105], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003480.pt"}
{"idx": 3482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTxn2BuqyzU.mp4", "duration": 176.04166666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy's stomach transform into a whole watermelon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy's stomach transform into a whole watermelon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 91.0], "base_se_norm": [0.011124253273010254, 0.033846139907836914], "gt_se_raw": [111.0, 118.0], "gt_se_norm": [0.2610650062561035, 0.3405916690826416], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003482.pt"}
{"idx": 3484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTxn2BuqyzU.mp4", "duration": 176.04166666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the watermelon boy fly off the swing and break into pieces on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the watermelon boy fly off the swing and break into pieces on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 151.0], "base_se_norm": [0.5791715383529663, 0.7155028581619263], "gt_se_raw": [144.0, 149.0], "gt_se_norm": [0.6359763145446777, 0.6927809715270996], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003484.pt"}
{"idx": 3486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J_dBkatloPk.mp4", "duration": 14.633999999999999, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the fans in the upper rows are most actively raising their arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the fans in the upper rows are most actively raising their arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.17999178171157837], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.8633319735527039, -0.45332783460617065], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003486.pt"}
{"idx": 3488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J_dBkatloPk.mp4", "duration": 14.633999999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point is a person in a white hockey jersey standing near the aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a person in a white hockey jersey standing near the aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.9133524894714355], "gt_se_raw": [6.0, 11.0], "gt_se_norm": [-0.17999178171157837, 0.5033483505249023], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003488.pt"}
{"idx": 3490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qwGfbIw206k.mp4", "duration": 47.48076666666667, "prompt_text": "Please find the visual event described by the sentence 'A woman in a pink top shows a child a pair of clothes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman in a pink top shows a child a pair of clothes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9578776955604553, -0.7051438093185425], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.7893884181976318, -0.7051438093185425], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003490.pt"}
{"idx": 3492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qwGfbIw206k.mp4", "duration": 47.48076666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the child reaches out to touch the pants on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the child reaches out to touch the pants on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.5366544723510742, -0.32604289054870605], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.558525800704956, 0.6427704095840454], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003492.pt"}
{"idx": 3494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qwGfbIw206k.mp4", "duration": 47.48076666666667, "prompt_text": "Please find the visual event described by the sentence 'A woman embraces a child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman embraces a child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [0.8112597465515137, 0.9376267194747925], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [0.8112597465515137, 0.9376267194747925], "start_token_span": [60, 64], "end_token_span": [66, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003494.pt"}
{"idx": 3496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVTOIJ4Q1DA.mp4", "duration": 97.66423333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman in the yellow top pointing at a machine on the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman in the yellow top pointing at a machine on the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 96.0], "base_se_norm": [0.8430494070053101, 0.9659193754196167], "gt_se_raw": [89.0, 94.0], "gt_se_norm": [0.8225710391998291, 0.9249626398086548], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003496.pt"}
{"idx": 3498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVTOIJ4Q1DA.mp4", "duration": 97.66423333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the women are looking at clothes on a rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the women are looking at clothes on a rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 87.0], "base_se_norm": [0.6177878379821777, 0.7816144227981567], "gt_se_raw": [75.0, 88.0], "gt_se_norm": [0.5358744859695435, 0.8020927906036377], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003498.pt"}
{"idx": 3500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XTW3LVp4pWA.mp4", "duration": 148.12, "prompt_text": "Please find the visual event described by the sentence 'When can we see the red banner with the text 'Diversity makes a world of difference'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the red banner with the text 'Diversity makes a world of difference'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.972994863986969], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9594923257827759], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003500.pt"}
{"idx": 3502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XTW3LVp4pWA.mp4", "duration": 148.12, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a close-up of an elevator's stainless steel doors is shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a close-up of an elevator's stainless steel doors is shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5274101495742798, -0.48690247535705566], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [-0.5139076709747314, -0.48690247535705566], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003502.pt"}
{"idx": 3504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XTW3LVp4pWA.mp4", "duration": 148.12, "prompt_text": "Please find the visual event described by the sentence 'At what point are people in academic gowns seen socializing in a large dining hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people in academic gowns seen socializing in a large dining hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 146.0], "base_se_norm": [0.8498514890670776, 0.9713746309280396], "gt_se_raw": [143.0, 146.0], "gt_se_norm": [0.9308669567108154, 0.9713746309280396], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003504.pt"}
{"idx": 3506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_gNYItV44A.mp4", "duration": 123.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the woman standing next to a large metal slide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman standing next to a large metal slide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 44.0], "base_se_norm": [-0.6433396339416504, -0.2866792678833008], "gt_se_raw": [22.0, 44.0], "gt_se_norm": [-0.6433396339416504, -0.2866792678833008], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003506.pt"}
{"idx": 3508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_gNYItV44A.mp4", "duration": 123.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the camera showing the swing sets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the camera showing the swing sets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.8703052997589111, -0.7568224668502808], "gt_se_raw": [8.0, 22.0], "gt_se_norm": [-0.8703052997589111, -0.6433396339416504], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003508.pt"}
{"idx": 3510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RF99-5G-Hrk.mp4", "duration": 102.73596666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a brown top holding up a black jacket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a brown top holding up a black jacket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 20.0], "base_se_norm": [-0.8247935771942139, -0.6106524467468262], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.6690545678138733, -0.5522502660751343], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003510.pt"}
{"idx": 3512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RF99-5G-Hrk.mp4", "duration": 102.73596666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a woman is holding a red quilted skirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a woman is holding a red quilted skirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 52.0], "base_se_norm": [-0.2018374800682068, 0.012303709983825684], "gt_se_raw": [40.0, 46.0], "gt_se_norm": [-0.22130483388900757, -0.10450059175491333], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003512.pt"}
{"idx": 3514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RF99-5G-Hrk.mp4", "duration": 102.73596666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the older woman is examining a blue jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the older woman is examining a blue jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 96.0], "base_se_norm": [0.6547272205352783, 0.868868350982666], "gt_se_raw": [86.0, 95.0], "gt_se_norm": [0.6741945743560791, 0.8494009971618652], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003514.pt"}
{"idx": 3516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KwOdedfBqHE.mp4", "duration": 196.82162499999998, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of the young woman with an orange beanie talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of the young woman with an orange beanie talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 51.0], "base_se_norm": [-0.6138635873794556, -0.4817642569541931], "gt_se_raw": [37.0, 52.0], "gt_se_norm": [-0.6240250468254089, -0.47160279750823975], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003516.pt"}
{"idx": 3518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KwOdedfBqHE.mp4", "duration": 196.82162499999998, "prompt_text": "Please find the visual event described by the sentence 'Find the shot that shows a water fountain in a pond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot that shows a water fountain in a pond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 121.0], "base_se_norm": [0.18889379501342773, 0.22953975200653076], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [0.19905519485473633, 0.22953975200653076], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003518.pt"}
{"idx": 3520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FY0f8PYebKo.mp4", "duration": 123.6235, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with the dog appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with the dog appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 16.0], "base_se_norm": [-0.9029310941696167, -0.7411495447158813], "gt_se_raw": [6.0, 16.0], "gt_se_norm": [-0.9029310941696167, -0.7411495447158813], "start_token_span": [69, 72], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003520.pt"}
{"idx": 3522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FY0f8PYebKo.mp4", "duration": 123.6235, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman wearing a red button-down shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman wearing a red button-down shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 37.0], "base_se_norm": [-0.514655351638794, -0.401408314704895], "gt_se_raw": [30.0, 38.0], "gt_se_norm": [-0.514655351638794, -0.3852301239967346], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003522.pt"}
{"idx": 3524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FY0f8PYebKo.mp4", "duration": 123.6235, "prompt_text": "Please find the visual event described by the sentence 'Locate the man in the striped cardigan and glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the man in the striped cardigan and glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 70.0], "base_se_norm": [-0.17491412162780762, 0.13247084617614746], "gt_se_raw": [51.0, 70.0], "gt_se_norm": [-0.17491412162780762, 0.13247084617614746], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003524.pt"}
{"idx": 3526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iGtZnKy7-jM.mp4", "duration": 19.4194, "prompt_text": "Please find the visual event described by the sentence 'When is the boy in the blue hoodie laughing while playing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy in the blue hoodie laughing while playing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.588040828704834, -0.382061243057251], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.6910306215286255, -0.382061243057251], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003526.pt"}
{"idx": 3528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iGtZnKy7-jM.mp4", "duration": 19.4194, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the drummer on the left crosses his sticks to hit a drum on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the drummer on the left crosses his sticks to hit a drum on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.02989792823791504, 0.23587751388549805], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [0.02989792823791504, 0.13288772106170654], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003528.pt"}
{"idx": 3530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZjejoT1gFOc.mp4", "duration": 198.63176666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a map of Europe shown to indicate the location of cave art sites?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a map of Europe shown to indicate the location of cave art sites?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 41.0], "base_se_norm": [-0.7482779026031494, -0.5871757864952087], "gt_se_raw": [24.0, 42.0], "gt_se_norm": [-0.7583467960357666, -0.5771068930625916], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003530.pt"}
{"idx": 3532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZjejoT1gFOc.mp4", "duration": 198.63176666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when a chart of different cave art symbols is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when a chart of different cave art symbols is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 128.0], "base_se_norm": [0.208266019821167, 0.28881704807281494], "gt_se_raw": [120.0, 128.0], "gt_se_norm": [0.208266019821167, 0.28881704807281494], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003532.pt"}
{"idx": 3534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZjejoT1gFOc.mp4", "duration": 198.63176666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that displays colorful paintings of bison heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that displays colorful paintings of bison heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 97.0], "base_se_norm": [-0.09380048513412476, -0.023318350315093994], "gt_se_raw": [89.0, 97.0], "gt_se_norm": [-0.10386937856674194, -0.023318350315093994], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003534.pt"}
{"idx": 3536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_vUG5rqC6qI.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When is a man using a laptop for vehicle diagnostics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man using a laptop for vehicle diagnostics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5414421558380127, -0.08288431167602539], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003536.pt"}
{"idx": 3538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_vUG5rqC6qI.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the text 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the text 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5066900253295898, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003538.pt"}
{"idx": 3540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b33M-dJu5gM.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher guiding the child to find letters on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher guiding the child to find letters on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 39.0], "base_se_norm": [-0.9626168012619019, 0.4579439163208008], "gt_se_raw": [7.0, 19.0], "gt_se_norm": [-0.7383177280426025, -0.2897196412086487], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003540.pt"}
{"idx": 3542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b33M-dJu5gM.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the teacher gives a thumbs-up gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the teacher gives a thumbs-up gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [-0.17757010459899902, -0.10280376672744751], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.17757010459899902, -0.10280376672744751], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003542.pt"}
{"idx": 3544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b33M-dJu5gM.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'When does the child touch his mouth and face while looking at the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child touch his mouth and face while looking at the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 15.0], "base_se_norm": [-0.8130841255187988, -0.4392523169517517], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.5887850522994995, -0.47663551568984985], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003544.pt"}
{"idx": 3546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AB2oAgjjt3g.mp4", "duration": 62.1621, "prompt_text": "Please find the visual event described by the sentence 'Show the shot of the 'GLOBAL ONLINE' sign in front of the building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the shot of the 'GLOBAL ONLINE' sign in front of the building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.6139127016067505, -0.4852169156074524], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.646086573600769, -0.5173908472061157], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003546.pt"}
{"idx": 3548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AB2oAgjjt3g.mp4", "duration": 62.1621, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman with blonde hair and glasses smiles and points at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman with blonde hair and glasses smiles and points at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [0.029566168785095215, 0.12608802318572998], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [0.029566168785095215, 0.12608802318572998], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003548.pt"}
{"idx": 3550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uDTB-0LqDw8.mp4", "duration": 152.5, "prompt_text": "Please find the visual event described by the sentence 'When can we see a cabbage white butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a cabbage white butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 76.0], "base_se_norm": [-0.16065573692321777, -0.003278672695159912], "gt_se_raw": [65.0, 76.0], "gt_se_norm": [-0.14754098653793335, -0.003278672695159912], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003550.pt"}
{"idx": 3552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uDTB-0LqDw8.mp4", "duration": 152.5, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the Tiger Swallowtail is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the Tiger Swallowtail is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 53.0], "base_se_norm": [-0.4622950553894043, -0.3049180507659912], "gt_se_raw": [42.0, 52.0], "gt_se_norm": [-0.4491803050041199, -0.31803280115127563], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003552.pt"}
{"idx": 3554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uDTB-0LqDw8.mp4", "duration": 152.5, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a specimen of the California dogface butterfly is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a specimen of the California dogface butterfly is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.6065573692321777, -0.55409836769104], "gt_se_raw": [84.0, 88.0], "gt_se_norm": [0.10163938999176025, 0.15409839153289795], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003554.pt"}
{"idx": 3556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s6Mk7Y3Q5QE.mp4", "duration": 38.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy rubs his face with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy rubs his face with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.273984432220459, -0.17026793956756592], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.273984432220459, -0.17026793956756592], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003556.pt"}
{"idx": 3558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s6Mk7Y3Q5QE.mp4", "duration": 38.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the blue shirt use hand gestures to explain something?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the blue shirt use hand gestures to explain something?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 37.0], "base_se_norm": [0.4001729488372803, 0.9187554121017456], "gt_se_raw": [28.0, 36.0], "gt_se_norm": [0.45203113555908203, 0.8668972253799438], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003558.pt"}
{"idx": 3560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PV34pW-53Os.mp4", "duration": 110.90245833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the elf in the green costume introduce himself and the video's topic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the elf in the green costume introduce himself and the video's topic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.7835935950279236, -0.6032549738883972], "gt_se_raw": [12.0, 22.0], "gt_se_norm": [-0.7835935950279236, -0.6032549738883972], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003560.pt"}
{"idx": 3562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PV34pW-53Os.mp4", "duration": 110.90245833333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does Santa mistakenly address the visually impaired adult as a 'little boy'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does Santa mistakenly address the visually impaired adult as a 'little boy'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 49.0], "base_se_norm": [-0.278645396232605, -0.11634057760238647], "gt_se_raw": [40.0, 52.0], "gt_se_norm": [-0.278645396232605, -0.062238991260528564], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003562.pt"}
{"idx": 3564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PV34pW-53Os.mp4", "duration": 110.90245833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the girl who was previously on Santa's lap receives a candy cane from an elf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the girl who was previously on Santa's lap receives a candy cane from an elf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 71.0], "base_se_norm": [0.20826900005340576, 0.2804044485092163], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [0.2263028621673584, 0.29843831062316895], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003564.pt"}
{"idx": 3566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Opbfh4wNu7Q.mp4", "duration": 60.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'What is the timeframe for the shot of a spaceship traveling through a wormhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the timeframe for the shot of a spaceship traveling through a wormhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 35.0], "base_se_norm": [0.09923660755157471, 0.16585707664489746], "gt_se_raw": [34.0, 35.0], "gt_se_norm": [0.1325467824935913, 0.16585707664489746], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003566.pt"}
{"idx": 3568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Opbfh4wNu7Q.mp4", "duration": 60.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man is lying on the ground next to a shield.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man is lying on the ground next to a shield.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 46.0], "base_se_norm": [0.4656487703323364, 0.5322692394256592], "gt_se_raw": [44.0, 45.0], "gt_se_norm": [0.4656487703323364, 0.4989590644836426], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003568.pt"}
{"idx": 3570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KhKaTh-YQNE.mp4", "duration": 55.522133333333336, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a purple shirt interacting with a boy in a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a purple shirt interacting with a boy in a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 55.0], "base_se_norm": [-1.0, 0.981191873550415], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.7838699817657471], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003570.pt"}
{"idx": 3572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KhKaTh-YQNE.mp4", "duration": 55.522133333333336, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the boy is left alone in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the boy is left alone in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 41.0], "base_se_norm": [-0.6397832632064819, 0.4768885374069214], "gt_se_raw": [11.0, 40.0], "gt_se_norm": [-0.603761613368988, 0.4408668279647827], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003572.pt"}
{"idx": 3574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KhKaTh-YQNE.mp4", "duration": 55.522133333333336, "prompt_text": "Please find the visual event described by the sentence 'When is the woman holding a small object on a string in front of the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman holding a small object on a string in front of the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [0.8731268644332886, 0.981191873550415], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [0.8731268644332886, 0.981191873550415], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003574.pt"}
{"idx": 3576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E07s5ZYygMg.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man wearing a colorful shirt lying on the sand, surrounded by women?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man wearing a colorful shirt lying on the sand, surrounded by women?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 77.0], "base_se_norm": [-0.3220828175544739, -0.1843808889389038], "gt_se_raw": [67.0, 72.0], "gt_se_norm": [-0.290305495262146, -0.2373431921005249], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003576.pt"}
{"idx": 3578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E07s5ZYygMg.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man is lying on the grass and a woman is holding his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man is lying on the grass and a woman is holding his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 98.0], "base_se_norm": [-0.004309177398681641, 0.038060665130615234], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.027468204498291016, 0.04865312576293945], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003578.pt"}
{"idx": 3580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tEyDRWmQEKs.mp4", "duration": 133.3332, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher giving instructions to the students who are sitting on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher giving instructions to the students who are sitting on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 34.0], "base_se_norm": [-0.9549999833106995, -0.48999953269958496], "gt_se_raw": [3.0, 35.0], "gt_se_norm": [-0.9549999833106995, -0.47499948740005493], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003580.pt"}
{"idx": 3582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tEyDRWmQEKs.mp4", "duration": 133.3332, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the student wearing glasses reflects on the feedback he received.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the student wearing glasses reflects on the feedback he received.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 131.0], "base_se_norm": [0.6950016021728516, 0.9650018215179443], "gt_se_raw": [94.0, 131.0], "gt_se_norm": [0.41000139713287354, 0.9650018215179443], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003582.pt"}
{"idx": 3584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tEyDRWmQEKs.mp4", "duration": 133.3332, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for when the text 'suggestions for improvement' is displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for when the text 'suggestions for improvement' is displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 104.0], "base_se_norm": [0.41000139713287354, 0.5600014925003052], "gt_se_raw": [94.0, 101.0], "gt_se_norm": [0.41000139713287354, 0.5150014162063599], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003584.pt"}
{"idx": 3586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hseWMRV3lA8.mp4", "duration": 119.753, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man with the mustache attempts to type but produces gibberish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man with the mustache attempts to type but produces gibberish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 87.0], "base_se_norm": [-0.6158760190010071, 0.4529907703399658], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.6325770616531372, -0.5490717887878418], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003586.pt"}
{"idx": 3588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hseWMRV3lA8.mp4", "duration": 119.753, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when the reporter on the instant messenger asks, 'Is everything OK?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when the reporter on the instant messenger asks, 'Is everything OK?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 67.0], "base_se_norm": [0.08556783199310303, 0.11896991729736328], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [0.08556783199310303, 0.11896991729736328], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003588.pt"}
{"idx": 3590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/II5UsqP2JAk.mp4", "duration": 181.0809, "prompt_text": "Please find the visual event described by the sentence 'When are two people riding horses through a snowy town?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two people riding horses through a snowy town?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 38.0], "base_se_norm": [-0.6686563491821289, -0.5802980661392212], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.6023876667022705, -0.5692533254623413], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003590.pt"}
{"idx": 3592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/II5UsqP2JAk.mp4", "duration": 181.0809, "prompt_text": "Please find the visual event described by the sentence 'A person on a horse arrives at a house in a snowy forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person on a horse arrives at a house in a snowy forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [-0.2820888161659241, -0.20477533340454102], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.2820888161659241, -0.24895447492599487], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003592.pt"}
{"idx": 3594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/II5UsqP2JAk.mp4", "duration": 181.0809, "prompt_text": "Please find the visual event described by the sentence 'A person in a boat approaches a broken-down ferris wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person in a boat approaches a broken-down ferris wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 111.0], "base_se_norm": [0.19283699989318848, 0.22597134113311768], "gt_se_raw": [109.0, 111.0], "gt_se_norm": [0.20388174057006836, 0.22597134113311768], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003594.pt"}
{"idx": 3596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B-dn8jMxKeU.mp4", "duration": 44.67796666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a woman in a pink hoodie is cooking in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a woman in a pink hoodie is cooking in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.37329286336898804, -0.2837632894515991], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.4180576801300049, -0.32852810621261597], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003596.pt"}
{"idx": 3598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B-dn8jMxKeU.mp4", "duration": 44.67796666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a person is holding a large pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a person is holding a large pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [0.34294378757476807, 0.43247342109680176], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [0.34294378757476807, 0.43247342109680176], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003598.pt"}
{"idx": 3600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PcnqtNANWVw.mp4", "duration": 471.9957224969924, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a police car driving on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a police car driving on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9957626461982727, -0.9872879981994629], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9957626461982727, -0.9872879981994629], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003600.pt"}
{"idx": 3602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PcnqtNANWVw.mp4", "duration": 471.9957224969924, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a man in a blue shirt points to a whiteboard with writing on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a man in a blue shirt points to a whiteboard with writing on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [-0.6737258434295654, -0.648301899433136], "gt_se_raw": [77.0, 83.0], "gt_se_norm": [-0.6737258434295654, -0.648301899433136], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003602.pt"}
{"idx": 3604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PcnqtNANWVw.mp4", "duration": 471.9957224969924, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man on a mobility scooter is talking to two other men.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 292.0 - 304.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man on a mobility scooter is talking to two other men.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 292.0 - 304.0 seconds.<|im_end|>\n", "base_se_raw": [292.0, 304.0], "base_se_norm": [0.23729932308197021, 0.2881472110748291], "gt_se_raw": [292.0, 298.0], "gt_se_norm": [0.23729932308197021, 0.26272332668304443], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003604.pt"}
{"idx": 3606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PUtHH2rw7KM.mp4", "duration": 96.36293333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the automated voice begin counting clearly for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the automated voice begin counting clearly for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 40.0], "base_se_norm": [-0.3773539066314697, -0.16980522871017456], "gt_se_raw": [40.0, 46.0], "gt_se_norm": [-0.16980522871017456, -0.04527604579925537], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003606.pt"}
{"idx": 3608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PUtHH2rw7KM.mp4", "duration": 96.36293333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the receiver is swept from the right, and the sound transitions from static to clear counting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the receiver is swept from the right, and the sound transitions from static to clear counting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 19.0], "base_se_norm": [-0.7716964483261108, -0.6056574583053589], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [0.20378243923187256, 0.34906649589538574], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003608.pt"}
{"idx": 3610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nhPG6tC_TKU.mp4", "duration": 50.38366666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of drilling a hole in the top of a coconut.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of drilling a hole in the top of a coconut.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [0.19086205959320068, 0.3496437072753906], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [0.19086205959320068, 0.3893390893936157], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003610.pt"}
{"idx": 3612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nhPG6tC_TKU.mp4", "duration": 50.38366666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the coconuts lined up on the shelves inside the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the coconuts lined up on the shelves inside the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.7618275880813599, -0.6030459403991699], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.7221322059631348, -0.642741322517395], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003612.pt"}
{"idx": 3614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nhPG6tC_TKU.mp4", "duration": 50.38366666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a person grabs a straw for their drink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a person grabs a straw for their drink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [0.3893390893936157, 0.5084253549575806], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.3893390893936157, 0.46872997283935547], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003614.pt"}
{"idx": 3616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2_Z1KdMtDD4.mp4", "duration": 29.963266666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are two people visible in the driver's cabin of the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two people visible in the driver's cabin of the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.13227081298828125, 0.13472270965576172], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.2657676339149475, 0.13472270965576172], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003616.pt"}
{"idx": 3618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2_Z1KdMtDD4.mp4", "duration": 29.963266666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the train is moving past the camera's viewpoint.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the train is moving past the camera's viewpoint.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 29.0], "base_se_norm": [-1.0, 0.9357035160064697], "gt_se_raw": [15.0, 28.0], "gt_se_norm": [0.0012259483337402344, 0.868955135345459], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003618.pt"}
{"idx": 3620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZLZOnqxa99c.mp4", "duration": 55.180125, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a large group of dolphins swimming in the deep blue ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a large group of dolphins swimming in the deep blue ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 12.0], "base_se_norm": [-0.9637550711631775, -0.5650607943534851], "gt_se_raw": [1.0, 12.0], "gt_se_norm": [-0.9637550711631775, -0.5650607943534851], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003620.pt"}
{"idx": 3622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZLZOnqxa99c.mp4", "duration": 55.180125, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when four dolphins are seen diving downwards in unison.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when four dolphins are seen diving downwards in unison.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.20261144638061523, 0.014858126640319824], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [-0.23885637521743774, 0.014858126640319824], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003622.pt"}
{"idx": 3624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZLZOnqxa99c.mp4", "duration": 55.180125, "prompt_text": "Please find the visual event described by the sentence 'Show the clip from a low angle where the dolphins are swimming up towards the water's surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip from a low angle where the dolphins are swimming up towards the water's surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 48.0], "base_se_norm": [-0.20261144638061523, 0.7397568225860596], "gt_se_raw": [40.0, 46.0], "gt_se_norm": [0.4497973918914795, 0.6672669649124146], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003624.pt"}
{"idx": 3626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gm5ItoIJ4sg.mp4", "duration": 49.5495, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the second knot is made, forming a small circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the second knot is made, forming a small circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 39.0], "base_se_norm": [0.08981931209564209, 0.574183464050293], "gt_se_raw": [17.0, 24.0], "gt_se_norm": [-0.31381750106811523, -0.03127175569534302], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003626.pt"}
{"idx": 3628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gm5ItoIJ4sg.mp4", "duration": 49.5495, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the second lace is inserted into the circle to form the second loop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the second lace is inserted into the circle to form the second loop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 39.0], "base_se_norm": [0.08981931209564209, 0.574183464050293], "gt_se_raw": [36.0, 41.0], "gt_se_norm": [0.4530923366546631, 0.6549108028411865], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003628.pt"}
{"idx": 3630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wvaG-frUcV4.mp4", "duration": 11.866666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see three hands interacting together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three hands interacting together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.685393214225769], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, 0.17977523803710938], "start_token_span": [69, 72], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003630.pt"}
{"idx": 3632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wvaG-frUcV4.mp4", "duration": 11.866666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a hand wearing a white beaded bracelet is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a hand wearing a white beaded bracelet is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.685393214225769], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.685393214225769], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003632.pt"}
{"idx": 3634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wvaG-frUcV4.mp4", "duration": 11.866666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the brief moment a hand touches the blue part of the playground equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the brief moment a hand touches the blue part of the playground equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 11.0], "base_se_norm": [0.685393214225769, 0.8539326190948486], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [0.685393214225769, 0.8539326190948486], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003634.pt"}
{"idx": 3636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x5U8zODIz9s.mp4", "duration": 142.20873333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the device says 'I want to fix train'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the device says 'I want to fix train'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 102.0], "base_se_norm": [0.307936429977417, 0.4345109462738037], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [-0.15617001056671143, -0.1280423402786255], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003636.pt"}
{"idx": 3638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x5U8zODIz9s.mp4", "duration": 142.20873333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where the woman points at the screen to show the location of a word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where the woman points at the screen to show the location of a word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 101.0], "base_se_norm": [0.293872594833374, 0.42044711112976074], "gt_se_raw": [92.0, 97.0], "gt_se_norm": [0.293872594833374, 0.36419177055358887], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003638.pt"}
{"idx": 3640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q-SGjT4CPtQ.mp4", "duration": 56.51479166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene with the bicycle parked by the tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene with the bicycle parked by the tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.7168883085250854, -0.5045545101165771], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.7168883085250854, -0.5045545101165771], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003640.pt"}
{"idx": 3642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q-SGjT4CPtQ.mp4", "duration": 56.51479166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that shows a building with large columns.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that shows a building with large columns.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [0.4155585765838623, 0.6986702680587769], "gt_se_raw": [40.0, 48.0], "gt_se_norm": [0.4155585765838623, 0.6986702680587769], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003642.pt"}
{"idx": 3644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q-SGjT4CPtQ.mp4", "duration": 56.51479166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the final logo screen for the College of Health and Human Sciences.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the final logo screen for the College of Health and Human Sciences.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [0.73405921459198, 0.9817819595336914], "gt_se_raw": [49.0, 53.0], "gt_se_norm": [0.73405921459198, 0.875615119934082], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003644.pt"}
{"idx": 3646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1zyNbdDgUb8.mp4", "duration": 466.3825833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip from the movie 'Cats'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip from the movie 'Cats'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [-0.9614050984382629, -0.9356751441955566], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.9614050984382629, -0.9356751441955566], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003646.pt"}
{"idx": 3648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1zyNbdDgUb8.mp4", "duration": 466.3825833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the cat is walking on a leash outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the cat is walking on a leash outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 61.0], "base_se_norm": [-0.7555654644966125, -0.7384121417999268], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [-0.7555654644966125, -0.7384121417999268], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003648.pt"}
{"idx": 3650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DAHbtsjuNws.mp4", "duration": 132.96616666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where Ginger Lane is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where Ginger Lane is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 24.0], "base_se_norm": [-0.9097514748573303, -0.6390058994293213], "gt_se_raw": [6.0, 13.0], "gt_se_norm": [-0.9097514748573303, -0.8044615387916565], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003650.pt"}
{"idx": 3652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DAHbtsjuNws.mp4", "duration": 132.96616666666668, "prompt_text": "Please find the visual event described by the sentence 'At what time do two women dance together, with one in a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do two women dance together, with one in a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 64.0], "base_se_norm": [-0.2629704475402832, -0.03734910488128662], "gt_se_raw": [50.0, 64.0], "gt_se_norm": [-0.24792903661727905, -0.03734910488128662], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003652.pt"}
{"idx": 3654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DAHbtsjuNws.mp4", "duration": 132.96616666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the wheelchair dance with other people on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the wheelchair dance with other people on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 98.0], "base_se_norm": [0.29356205463409424, 0.4740591049194336], "gt_se_raw": [86.0, 99.0], "gt_se_norm": [0.29356205463409424, 0.4891005754470825], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003654.pt"}
{"idx": 3656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/emx92kBKads.mp4", "duration": 315.76666666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a boy is writing on a large green chalkboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a boy is writing on a large green chalkboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 59.0], "base_se_norm": [-0.6516414880752563, -0.6263062953948975], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.6516414880752563, -0.6326401233673096], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003656.pt"}
{"idx": 3658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/emx92kBKads.mp4", "duration": 315.76666666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the information about the number of searches on Google per month displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the information about the number of searches on Google per month displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 132.0], "base_se_norm": [-0.23361128568649292, -0.16393959522247314], "gt_se_raw": [121.0, 127.0], "gt_se_norm": [-0.23361128568649292, -0.19560855627059937], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003658.pt"}
{"idx": 3660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wKcPU0WPdqE.mp4", "duration": 89.82315648982316, "prompt_text": "Please find the visual event described by the sentence 'When can we see the three men running on the sidewalk away from the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the three men running on the sidewalk away from the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 29.0], "base_se_norm": [-0.5101485848426819, -0.354286789894104], "gt_se_raw": [22.0, 29.0], "gt_se_norm": [-0.5101485848426819, -0.354286789894104], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003660.pt"}
{"idx": 3662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wKcPU0WPdqE.mp4", "duration": 89.82315648982316, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the blue tank top speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the blue tank top speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 60.0], "base_se_norm": [0.22462844848632812, 0.3359583616256714], "gt_se_raw": [54.0, 59.0], "gt_se_norm": [0.2023625373840332, 0.31369245052337646], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003662.pt"}
{"idx": 3664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wKcPU0WPdqE.mp4", "duration": 89.82315648982316, "prompt_text": "Please find the visual event described by the sentence 'When are the men running along the coast?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the men running along the coast?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 67.0], "base_se_norm": [0.13556456565856934, 0.49182021617889404], "gt_se_raw": [64.0, 68.0], "gt_se_norm": [0.4250222444534302, 0.514086127281189], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003664.pt"}
{"idx": 3666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aqXFcRRQZCM.mp4", "duration": 56.181124999999994, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the bald man looks down in disappointment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the bald man looks down in disappointment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 24.0], "base_se_norm": [-0.39481455087661743, -0.14562052488327026], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.1812196969985962, -0.11002141237258911], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003666.pt"}
{"idx": 3668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aqXFcRRQZCM.mp4", "duration": 56.181124999999994, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the floral dress touch her necklace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the floral dress touch her necklace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [0.032375216484069824, 0.13917267322540283], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.032375216484069824, 0.13917267322540283], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003668.pt"}
{"idx": 3670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wekSrZ-d1bM.mp4", "duration": 113.94728061394729, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman playing the piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman playing the piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 19.0], "base_se_norm": [-0.7893762588500977, -0.6665124893188477], "gt_se_raw": [11.0, 19.0], "gt_se_norm": [-0.806928277015686, -0.6665124893188477], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003670.pt"}
{"idx": 3672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wekSrZ-d1bM.mp4", "duration": 113.94728061394729, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the news article about remote learning resources?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the news article about remote learning resources?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.6489604711532593, -0.5963045358657837], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.6489604711532593, -0.5963045358657837], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003672.pt"}
{"idx": 3674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wekSrZ-d1bM.mp4", "duration": 113.94728061394729, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with students playing trombones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with students playing trombones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 61.0], "base_se_norm": [-0.15750515460968018, 0.07067060470581055], "gt_se_raw": [48.0, 61.0], "gt_se_norm": [-0.15750515460968018, 0.07067060470581055], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003674.pt"}
{"idx": 3676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YMpCN8CmZh8.mp4", "duration": 73.20646666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman is pointing at the light switch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman is pointing at the light switch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 39.0], "base_se_norm": [-0.09844028949737549, 0.06547963619232178], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.07112032175064087, 0.03815960884094238], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003676.pt"}
{"idx": 3678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YMpCN8CmZh8.mp4", "duration": 73.20646666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman opens the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman opens the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [0.09279966354370117, 0.3113595247268677], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [0.8577593564987183, 0.9397193193435669], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003678.pt"}
{"idx": 3680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdn2xstSvwo.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a green cap operating a large industrial machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a green cap operating a large industrial machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.540437638759613], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003680.pt"}
{"idx": 3682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdn2xstSvwo.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is a person designing a 3D model on a computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a person designing a 3D model on a computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.21217882633209229, -0.08087527751922607], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.21217882633209229, -0.08087527751922607], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003682.pt"}
{"idx": 3684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdn2xstSvwo.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the man wearing safety glasses while working on a lathe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the man wearing safety glasses while working on a lathe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 28.0], "base_se_norm": [0.3786870241165161, 0.8382494449615479], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4443387985229492, 0.8382494449615479], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003684.pt"}
{"idx": 3686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODAfdbHBtGE.mp4", "duration": 37.971266666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman in the colorful cardigan enters and starts talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman in the colorful cardigan enters and starts talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 10.0], "base_se_norm": [-0.8946571946144104, -0.473285973072052], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.841985821723938, -0.6839715838432312], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003686.pt"}
{"idx": 3688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODAfdbHBtGE.mp4", "duration": 37.971266666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the woman puts her hand on her chest to emphasize her point.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the woman puts her hand on her chest to emphasize her point.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 28.0], "base_se_norm": [0.0007567405700683594, 0.4747992753982544], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.0007567405700683594, 0.15877091884613037], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003688.pt"}
{"idx": 3690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H3DvNMAEWyM.mp4", "duration": 16.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the blue shirt demonstrate the sign for laughter by shaking the other woman's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the blue shirt demonstrate the sign for laughter by shaking the other woman's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.2756539583206177, 0.2072434425354004], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.2756539583206177, 0.08651900291442871], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003690.pt"}
{"idx": 3692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H3DvNMAEWyM.mp4", "duration": 16.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman in the black shirt turns her back to the other woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman in the black shirt turns her back to the other woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [0.08651900291442871, 0.44869208335876465], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [0.2072434425354004, 0.3279677629470825], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003692.pt"}
{"idx": 3694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H3DvNMAEWyM.mp4", "duration": 16.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the blue shirt first place her hand on the other woman's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the blue shirt first place her hand on the other woman's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.6378269791603088, -0.3963782787322998], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8792756795883179], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003694.pt"}
{"idx": 3696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qlvtkpn9Xn8.mp4", "duration": 127.375, "prompt_text": "Please find the visual event described by the sentence 'When does the interpreter move the black chair so he can sit next to the patient?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the interpreter move the black chair so he can sit next to the patient?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.6388616561889648, -0.5603532791137695], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.6388616561889648, -0.5603532791137695], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003696.pt"}
{"idx": 3698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qlvtkpn9Xn8.mp4", "duration": 127.375, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the patient interrupts the doctor and asks him to look at her when he speaks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the patient interrupts the doctor and asks him to look at her when he speaks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 74.0], "base_se_norm": [0.03631007671356201, 0.16192340850830078], "gt_se_raw": [70.0, 76.0], "gt_se_norm": [0.09911680221557617, 0.19332683086395264], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003698.pt"}
{"idx": 3700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1OJ14e6EMQ.mp4", "duration": 7.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy on his knees on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy on his knees on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.24369746446609497], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.49579834938049316], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003700.pt"}
{"idx": 3702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1OJ14e6EMQ.mp4", "duration": 7.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child in pajamas touch his shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child in pajamas touch his shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [0.008403301239013672, 0.5126050710678101], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [0.008403301239013672, 0.26050424575805664], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003702.pt"}
{"idx": 3704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1OJ14e6EMQ.mp4", "duration": 7.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the boy is sitting on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the boy is sitting on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.24369746446609497, 0.7647058963775635], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.49579834938049316, 0.7647058963775635], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003704.pt"}
{"idx": 3706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQz8oCsHbY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where Cortana sings 'You are my sunshine'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where Cortana sings 'You are my sunshine'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 30.0], "base_se_norm": [-0.8509777784347534, -0.7516295909881592], "gt_se_raw": [19.0, 28.0], "gt_se_norm": [-0.8426987528800964, -0.7681876420974731], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003706.pt"}
{"idx": 3708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQz8oCsHbY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Cortana state that it cannot create reminders in the past?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Cortana state that it cannot create reminders in the past?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 102.0], "base_se_norm": [-0.21349364519119263, -0.1555405855178833], "gt_se_raw": [104.0, 108.0], "gt_se_norm": [-0.13898253440856934, -0.10586649179458618], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003708.pt"}
{"idx": 3710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VsJKPdOuzZw.mp4", "duration": 93.21812499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the man reveal that he has had a botox injection?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man reveal that he has had a botox injection?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.35166847705841064, 0.4589437246322632], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.8283595442771912, -0.7854494452476501], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003710.pt"}
{"idx": 3712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VsJKPdOuzZw.mp4", "duration": 93.21812499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman express her frustration by saying 'I can't live like this anymore'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman express her frustration by saying 'I can't live like this anymore'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.2658482789993286, 0.35166847705841064], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.20616298913955688, -0.1632528305053711], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003712.pt"}
{"idx": 3714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VsJKPdOuzZw.mp4", "duration": 93.21812499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the woman mention turning their lawn into an ice rink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman mention turning their lawn into an ice rink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.5923539400100708, -0.5279887914657593], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [0.39457857608795166, 0.4374886751174927], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003714.pt"}
{"idx": 3716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T8NJLxTIEO0.mp4", "duration": 91.9919, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman's hand interacting with a weather application on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman's hand interacting with a weather application on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [-0.5434380769729614, -0.499955952167511], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.5434380769729614, -0.499955952167511], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003716.pt"}
{"idx": 3718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T8NJLxTIEO0.mp4", "duration": 91.9919, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where several people are standing in line at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where several people are standing in line at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [-0.3695096969604492, -0.3260275721549988], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.3695096969604492, -0.3260275721549988], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003718.pt"}
{"idx": 3720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3eMgzKC39Q.mp4", "duration": 196.70000000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the blue player get trapped in the hole by the red player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue player get trapped in the hole by the red player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 31.0], "base_se_norm": [-0.7763091325759888, -0.6847991943359375], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.7559735774993896, -0.7051347494125366], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003720.pt"}
{"idx": 3722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3eMgzKC39Q.mp4", "duration": 196.70000000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the red player is using the Item Shop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the red player is using the Item Shop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 138.0], "base_se_norm": [0.35231316089630127, 0.4031519889831543], "gt_se_raw": [133.0, 139.0], "gt_se_norm": [0.35231316089630127, 0.41331982612609863], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003722.pt"}
{"idx": 3724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3eMgzKC39Q.mp4", "duration": 196.70000000000002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the blue player falls off the bridge after being rescued.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 168.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the blue player falls off the bridge after being rescued.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 168.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [168.0, 173.0], "base_se_norm": [0.708185076713562, 0.759023904800415], "gt_se_raw": [170.0, 173.0], "gt_se_norm": [0.7285206317901611, 0.759023904800415], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003724.pt"}
{"idx": 3726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eTVqkWp7oCs.mp4", "duration": 57.265541666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the website is displayed on a computer monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the website is displayed on a computer monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 21.0], "base_se_norm": [-0.5459747910499573, -0.2665746212005615], "gt_se_raw": [13.0, 21.0], "gt_se_norm": [-0.5459747910499573, -0.2665746212005615], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003726.pt"}
{"idx": 3728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eTVqkWp7oCs.mp4", "duration": 57.265541666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the diagram with 'Labour Market Information' at the center is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the diagram with 'Labour Market Information' at the center is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 44.0], "base_se_norm": [0.25730061531066895, 0.5367007255554199], "gt_se_raw": [36.0, 44.0], "gt_se_norm": [0.25730061531066895, 0.5367007255554199], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003728.pt"}
{"idx": 3730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rdgoehu_0SE.mp4", "duration": 346.0790666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman with short dark hair, wearing a grey vest over a black top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman with short dark hair, wearing a grey vest over a black top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.924872636795044], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.924872636795044], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003730.pt"}
{"idx": 3732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rdgoehu_0SE.mp4", "duration": 346.0790666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman with red hair speaks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman with red hair speaks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 99.0], "base_se_norm": [-0.5781310796737671, -0.42787641286849976], "gt_se_raw": [73.0, 99.0], "gt_se_norm": [-0.5781310796737671, -0.42787641286849976], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003732.pt"}
{"idx": 3734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rdgoehu_0SE.mp4", "duration": 346.0790666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a man with dreadlocks and a red tie is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 309.0 - 316.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a man with dreadlocks and a red tie is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 309.0 - 316.0 seconds.<|im_end|>\n", "base_se_raw": [309.0, 316.0], "base_se_norm": [0.7857190370559692, 0.8261722326278687], "gt_se_raw": [308.0, 316.0], "gt_se_norm": [0.7799400091171265, 0.8261722326278687], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003734.pt"}
{"idx": 3736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uE74-8YAV9E.mp4", "duration": 122.56, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where an analogy is made between a house's foundation and minor supporting details.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where an analogy is made between a house's foundation and minor supporting details.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 58.0], "base_se_norm": [-0.2167101502418518, -0.053524792194366455], "gt_se_raw": [48.0, 58.0], "gt_se_norm": [-0.2167101502418518, -0.053524792194366455], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003736.pt"}
{"idx": 3738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uE74-8YAV9E.mp4", "duration": 122.56, "prompt_text": "Please find the visual event described by the sentence 'When does the presenter give an example about dogs making the best pets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the presenter give an example about dogs making the best pets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 96.0], "base_se_norm": [0.3054831027984619, 0.5665796995162964], "gt_se_raw": [80.0, 96.0], "gt_se_norm": [0.3054831027984619, 0.5665796995162964], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003738.pt"}
{"idx": 3740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lO_ibzMt9MQ.mp4", "duration": 36.54496168133596, "prompt_text": "Please find the visual event described by the sentence 'When is the woman introducing herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman introducing herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.781091570854187], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9452729225158691, -0.8358186483383179], "start_token_span": [61, 64], "end_token_span": [66, 69], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003740.pt"}
{"idx": 3742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lO_ibzMt9MQ.mp4", "duration": 36.54496168133596, "prompt_text": "Please find the visual event described by the sentence 'At what point is the rat running from the right to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the rat running from the right to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.06963914632797241, 0.09454214572906494], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.03981506824493408, 0.2039964199066162], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003742.pt"}
{"idx": 3744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lO_ibzMt9MQ.mp4", "duration": 36.54496168133596, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman is holding a red pouch and looking towards the rat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman is holding a red pouch and looking towards the rat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 26.0], "base_se_norm": [-0.45272892713546753, 0.4229048490524292], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [0.36817777156829834, 0.5323590040206909], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003744.pt"}
{"idx": 3746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BWYNt1N6xWQ.mp4", "duration": 259.7595, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the teacher spells the letter 'C' into the girl's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the teacher spells the letter 'C' into the girl's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 76.0], "base_se_norm": [-0.6920228004455566, -0.4148433208465576], "gt_se_raw": [41.0, 48.0], "gt_se_norm": [-0.6843233704566956, -0.630427360534668], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003746.pt"}
{"idx": 3748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BWYNt1N6xWQ.mp4", "duration": 259.7595, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl throws her doll on the ground in anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 153.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl throws her doll on the ground in anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 153.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [153.0, 155.0], "base_se_norm": [0.17801272869110107, 0.19341158866882324], "gt_se_raw": [152.0, 155.0], "gt_se_norm": [0.17031335830688477, 0.19341158866882324], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003748.pt"}
{"idx": 3750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lN05UfmIAh4.mp4", "duration": 61.04, "prompt_text": "Please find the visual event described by the sentence 'When does the general manager, Lloyd, enter the interview room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the general manager, Lloyd, enter the interview room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [0.17955434322357178, 0.40891218185424805], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [0.2123197317123413, 0.31061601638793945], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003750.pt"}
{"idx": 3752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lN05UfmIAh4.mp4", "duration": 61.04, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the manager leans in and asks the applicant a surprising question.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the manager leans in and asks the applicant a surprising question.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 53.0], "base_se_norm": [0.343381404876709, 0.7365661859512329], "gt_se_raw": [49.0, 53.0], "gt_se_norm": [0.6055045127868652, 0.7365661859512329], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003752.pt"}
{"idx": 3754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lN05UfmIAh4.mp4", "duration": 61.04, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the female interviewer is explaining the need to get approval from the general manager.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the female interviewer is explaining the need to get approval from the general manager.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 28.0], "base_se_norm": [-0.6068152189254761, -0.08256882429122925], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.6395806074142456, -0.4102228283882141], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003754.pt"}
{"idx": 3756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mv1FKi_-A1I.mp4", "duration": 138.27146666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are they working with the red-patterned paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are they working with the red-patterned paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 40.0], "base_se_norm": [-0.6817854046821594, -0.42142802476882935], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.6817854046821594, -0.6094639301300049], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003756.pt"}
{"idx": 3758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mv1FKi_-A1I.mp4", "duration": 138.27146666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the brown cardboard base being folded into shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the brown cardboard base being folded into shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 40.0], "base_se_norm": [-0.6817854046821594, -0.42142802476882935], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.5949996113777161, -0.5371423959732056], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003758.pt"}
{"idx": 3760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FldgzyEOSH8.mp4", "duration": 491.79130000000004, "prompt_text": "Please find the visual event described by the sentence 'When is the college building shown with green lighting at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the college building shown with green lighting at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 51.0], "base_se_norm": [-0.8251290917396545, -0.7925949096679688], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [-0.8210623264312744, -0.7925949096679688], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003760.pt"}
{"idx": 3762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FldgzyEOSH8.mp4", "duration": 491.79130000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point are people taking a large group photo, with many giving a thumbs-up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people taking a large group photo, with many giving a thumbs-up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 124.0], "base_se_norm": [-0.5241883993148804, -0.4957210421562195], "gt_se_raw": [118.0, 124.0], "gt_se_norm": [-0.5201216340065002, -0.4957210421562195], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003762.pt"}
{"idx": 3764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FldgzyEOSH8.mp4", "duration": 491.79130000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of students and an instructor working on a LEGO robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 376.0 - 380.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of students and an instructor working on a LEGO robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 376.0 - 380.0 seconds.<|im_end|>\n", "base_se_raw": [376.0, 380.0], "base_se_norm": [0.5291039943695068, 0.5453710556030273], "gt_se_raw": [376.0, 380.0], "gt_se_norm": [0.5291039943695068, 0.5453710556030273], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003764.pt"}
{"idx": 3766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ynUw0YsrmSg.mp4", "duration": 218.13458333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where they are doing plank with leg raises.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where they are doing plank with leg raises.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 82.0], "base_se_norm": [-0.514061450958252, -0.24817055463790894], "gt_se_raw": [52.0, 61.0], "gt_se_norm": [-0.5232300758361816, -0.4407122731208801], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003766.pt"}
{"idx": 3768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ynUw0YsrmSg.mp4", "duration": 218.13458333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment they perform a side plank on their left arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment they perform a side plank on their left arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 138.0], "base_se_norm": [0.045226335525512695, 0.26527392864227295], "gt_se_raw": [92.0, 118.0], "gt_se_norm": [-0.15648406744003296, 0.081900954246521], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003768.pt"}
{"idx": 3770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JNyn_w3hdZ0.mp4", "duration": 145.56208333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person holding a blue sign that reads 'HOME OF A UWF TOP SCHOLAR'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person holding a blue sign that reads 'HOME OF A UWF TOP SCHOLAR'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.8900812864303589, -0.8213820457458496], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8763414025306702, -0.8213820457458496], "start_token_span": [78, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003770.pt"}
{"idx": 3772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JNyn_w3hdZ0.mp4", "duration": 145.56208333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a woman in a blue patterned dress gives balloons to a girl in a red dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a woman in a blue patterned dress gives balloons to a girl in a red dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.34048759937286377, -0.27178841829299927], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.3267477750778198, -0.2855282425880432], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003772.pt"}
{"idx": 3774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JNyn_w3hdZ0.mp4", "duration": 145.56208333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time a young woman in a grey t-shirt is holding a large 'U' balloon and smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time a young woman in a grey t-shirt is holding a large 'U' balloon and smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 139.0], "base_se_norm": [0.8548785448074341, 0.9098379611968994], "gt_se_raw": [135.0, 139.0], "gt_se_norm": [0.8548785448074341, 0.9098379611968994], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003774.pt"}
{"idx": 3776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8bRkdkfWv3g.mp4", "duration": 110.92, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a man carrying a large bunch of bananas on his shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a man carrying a large bunch of bananas on his shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.5311936140060425, -0.4590696096420288], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.5131626129150391, -0.4590696096420288], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003776.pt"}
{"idx": 3778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8bRkdkfWv3g.mp4", "duration": 110.92, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the fruits get thrown into the trash can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the fruits get thrown into the trash can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [0.4605121612548828, 0.5146051645278931], "gt_se_raw": [81.0, 86.0], "gt_se_norm": [0.4605121612548828, 0.5506671667098999], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003778.pt"}
{"idx": 3780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ansWZq7yULE.mp4", "duration": 63.6636, "prompt_text": "Please find the visual event described by the sentence 'When can we see a large group of turtles swimming together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a large group of turtles swimming together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9057546257972717, -0.7172638773918152], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.9371697306632996, -0.7800941467285156], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003780.pt"}
{"idx": 3782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ansWZq7yULE.mp4", "duration": 63.6636, "prompt_text": "Please find the visual event described by the sentence 'At what point does the big turtle introduce the clownfish to his son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the big turtle introduce the clownfish to his son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 33.0], "base_se_norm": [-0.27745211124420166, 0.03669917583465576], "gt_se_raw": [20.0, 24.0], "gt_se_norm": [-0.37169748544692993, -0.24603700637817383], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003782.pt"}
{"idx": 3784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ansWZq7yULE.mp4", "duration": 63.6636, "prompt_text": "Please find the visual event described by the sentence 'When does the blue fish show up in the middle of all the turtles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue fish show up in the middle of all the turtles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [0.8220772743225098, 0.9477378129959106], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [0.8220772743225098, 0.9163225889205933], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003784.pt"}
{"idx": 3786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HZ21eT9lyog.mp4", "duration": 31.36466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the time range where a woman in a plaid shirt is stirring a pot on the stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time range where a woman in a plaid shirt is stirring a pot on the stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.6174038648605347, -0.4898717999458313], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.6174038648605347, -0.5536378622055054], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003786.pt"}
{"idx": 3788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HZ21eT9lyog.mp4", "duration": 31.36466666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the participants coating pork chops in a bowl of breading?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the participants coating pork chops in a bowl of breading?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [0.5303845405578613, 0.9129806756973267], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.7854486703872681, 0.9129806756973267], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003788.pt"}
{"idx": 3790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FwqsdGLhdgA.mp4", "duration": 65.44037499999999, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman is humming while using the vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman is humming while using the vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, -0.5110052824020386], "gt_se_raw": [0.0, 15.0], "gt_se_norm": [-1.0, -0.5415674448013306], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003790.pt"}
{"idx": 3792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FwqsdGLhdgA.mp4", "duration": 65.44037499999999, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where the woman is holding and looking at the paper that says 'DAD MAGAZINE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where the woman is holding and looking at the paper that says 'DAD MAGAZINE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 30.0], "base_se_norm": [-0.4193187355995178, -0.08313482999801636], "gt_se_raw": [20.0, 30.0], "gt_se_norm": [-0.38875657320022583, -0.08313482999801636], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003792.pt"}
{"idx": 3794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FwqsdGLhdgA.mp4", "duration": 65.44037499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the woman turns off the vacuum cleaner for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the woman turns off the vacuum cleaner for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [-0.5415674448013306, -0.4804430603981018], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.5415674448013306, -0.4804430603981018], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003794.pt"}
{"idx": 3796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQm5doF_Uc.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman trying to fix the snowman's nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman trying to fix the snowman's nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 94.0], "base_se_norm": [0.1676422357559204, 0.23324012756347656], "gt_se_raw": [88.0, 90.0], "gt_se_norm": [0.15452265739440918, 0.18076181411743164], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003796.pt"}
{"idx": 3798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQm5doF_Uc.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the snowman is melting near the fireplace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the snowman is melting near the fireplace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 122.0], "base_se_norm": [0.5218707323074341, 0.6005882024765015], "gt_se_raw": [117.0, 122.0], "gt_se_norm": [0.5349903106689453, 0.6005882024765015], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003798.pt"}
{"idx": 3800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r5L7Iokg5RY.mp4", "duration": 25.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the sequence where a DNA helix is being analyzed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the sequence where a DNA helix is being analyzed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 8.0], "base_se_norm": [-0.8400000333786011, -0.36000001430511475], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.8400000333786011, -0.6800000071525574], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003800.pt"}
{"idx": 3802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r5L7Iokg5RY.mp4", "duration": 25.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where two dinosaurs are seen biting each other's heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where two dinosaurs are seen biting each other's heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [0.039999961853027344, 0.2799999713897705], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.04000002145767212, 0.2799999713897705], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003802.pt"}
{"idx": 3804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r5L7Iokg5RY.mp4", "duration": 25.0, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the 'Jurassic World Evolution' title screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the 'Jurassic World Evolution' title screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [0.7599999904632568, 0.9199999570846558], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [0.6799999475479126, 1.0], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003804.pt"}
{"idx": 3806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZP7sEgR4Jic.mp4", "duration": 199.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the player use a skeleton egg and then fight the skeleton?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the player use a skeleton egg and then fight the skeleton?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 118.0], "base_se_norm": [0.04173624515533447, 0.18196988105773926], "gt_se_raw": [105.0, 118.0], "gt_se_norm": [0.05175292491912842, 0.18196988105773926], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003806.pt"}
{"idx": 3808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZP7sEgR4Jic.mp4", "duration": 199.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the player successfully plants the sapling on the dirt block in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the player successfully plants the sapling on the dirt block in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 69.0], "base_se_norm": [-0.37896496057510376, -0.3088480830192566], "gt_se_raw": [88.0, 91.0], "gt_se_norm": [-0.11853092908859253, -0.08848083019256592], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003808.pt"}
{"idx": 3810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9qsy9kvwPk.mp4", "duration": 52.986266666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the patterned sweater bend over with his head on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the patterned sweater bend over with his head on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8490175008773804], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8490175008773804], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003810.pt"}
{"idx": 3812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9qsy9kvwPk.mp4", "duration": 52.986266666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy in the patterned sweater is sitting on the chair and puts the toy on his leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy in the patterned sweater is sitting on the chair and puts the toy on his leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.6225436925888062, -0.4715611934661865], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.8112718462944031, -0.6225436925888062], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003812.pt"}
{"idx": 3814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RN2P75hH74.mp4", "duration": 62.97958333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the three people sitting at the table and eating silently?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three people sitting at the table and eating silently?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 28.0], "base_se_norm": [-1.0, -0.11082297563552856], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.5871677994728088], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003814.pt"}
{"idx": 3816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RN2P75hH74.mp4", "duration": 62.97958333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman in the floral dress reveals the 'DAD MAGAZINE'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman in the floral dress reveals the 'DAD MAGAZINE'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [-0.07906663417816162, 0.11147129535675049], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.07906663417816162, -0.01555401086807251], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003816.pt"}
{"idx": 3818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RN2P75hH74.mp4", "duration": 62.97958333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the family's conversation at the dinner table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the family's conversation at the dinner table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 43.0], "base_se_norm": [-1.0, 0.3655219078063965], "gt_se_raw": [14.0, 23.0], "gt_se_norm": [-0.5554114580154419, -0.26960456371307373], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003818.pt"}
{"idx": 3820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qX9FSZJu448.mp4", "duration": 294.12716666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the man wearing a tie-dye shirt while attempting a headstand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man wearing a tie-dye shirt while attempting a headstand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 188.0], "base_se_norm": [0.18996143341064453, 0.2783585786819458], "gt_se_raw": [175.0, 188.0], "gt_se_norm": [0.18996143341064453, 0.2783585786819458], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003820.pt"}
{"idx": 3822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qX9FSZJu448.mp4", "duration": 294.12716666666665, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man, wearing a red shirt, starts to walk without crutches.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 198.0 - 208.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man, wearing a red shirt, starts to walk without crutches.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 198.0 - 208.0 seconds.<|im_end|>\n", "base_se_raw": [198.0, 208.0], "base_se_norm": [0.3463563919067383, 0.41435420513153076], "gt_se_raw": [197.0, 205.0], "gt_se_norm": [0.3395566940307617, 0.393954873085022], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003822.pt"}
{"idx": 3824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GR2o6k8aPlI.mp4", "duration": 189.75, "prompt_text": "Please find the visual event described by the sentence 'When does the baby shark appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby shark appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8629776239395142, -0.8313570618629456], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.8629776239395142, -0.8313570618629456], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003824.pt"}
{"idx": 3826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GR2o6k8aPlI.mp4", "duration": 189.75, "prompt_text": "Please find the visual event described by the sentence 'When do all the sharks swim together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the sharks swim together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 68.0], "base_se_norm": [-0.4624505639076233, -0.28326743841171265], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.31488800048828125, -0.28326743841171265], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003826.pt"}
{"idx": 3828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GR2o6k8aPlI.mp4", "duration": 189.75, "prompt_text": "Please find the visual event described by the sentence 'When does the whale appear to save the little fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 150.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the whale appear to save the little fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 150.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [150.0, 161.0], "base_se_norm": [0.581027626991272, 0.696969747543335], "gt_se_raw": [150.0, 154.0], "gt_se_norm": [0.581027626991272, 0.6231883764266968], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003828.pt"}
{"idx": 3830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IDIQKPApIxM.mp4", "duration": 55.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl give a thumbs-up gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl give a thumbs-up gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.38219261169433594, -0.27316778898239136], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.41853421926498413, -0.34585100412368774], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003830.pt"}
{"idx": 3832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IDIQKPApIxM.mp4", "duration": 55.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the girl seen holding a long tool that looks like a cane with a marker at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl seen holding a long tool that looks like a cane with a marker at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 54.0], "base_se_norm": [0.3446396589279175, 0.9624470472335815], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [0.3082979917526245, 0.41732287406921387], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003832.pt"}
{"idx": 3834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0hhbMi5xFtc.mp4", "duration": 163.83033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the man poking holes in the water bottle with a pin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man poking holes in the water bottle with a pin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 40.0], "base_se_norm": [-0.6459752321243286, -0.5116899609565735], "gt_se_raw": [30.0, 39.0], "gt_se_norm": [-0.6337674856185913, -0.5238977074623108], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003834.pt"}
{"idx": 3836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0hhbMi5xFtc.mp4", "duration": 163.83033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the prank is demonstrated on a person in an office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the prank is demonstrated on a person in an office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 81.0], "base_se_norm": [-0.0844186544418335, -0.011172175407409668], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [-0.07221090793609619, -0.011172175407409668], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003836.pt"}
{"idx": 3838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0hhbMi5xFtc.mp4", "duration": 163.83033333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the orange juice spraying out of the bottle after the cap is removed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the orange juice spraying out of the bottle after the cap is removed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 124.0], "base_se_norm": [0.4405146837234497, 0.5137611627578735], "gt_se_raw": [119.0, 126.0], "gt_se_norm": [0.452722430229187, 0.5381766557693481], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003838.pt"}
{"idx": 3840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkQyRE0byBI.mp4", "duration": 155.86404166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman, looking scared, hides behind a large bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman, looking scared, hides behind a large bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.7690294981002808, -0.7048709988594055], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.7561978101730347, -0.7048709988594055], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003840.pt"}
{"idx": 3842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkQyRE0byBI.mp4", "duration": 155.86404166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man grabs the woman and forcefully kisses her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man grabs the woman and forcefully kisses her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 84.0], "base_se_norm": [-0.0632861852645874, 0.07786250114440918], "gt_se_raw": [69.0, 82.0], "gt_se_norm": [-0.11461299657821655, 0.05219900608062744], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003842.pt"}
{"idx": 3844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-8qvGAWiPQY.mp4", "duration": 35.32695833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a plaid shirt making a handicraft?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a plaid shirt making a handicraft?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9433860182762146, -0.8301580548286438], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.8867720365524292, -0.8301580548286438], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003844.pt"}
{"idx": 3846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-8qvGAWiPQY.mp4", "duration": 35.32695833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point can we observe two women shaking hands at the reception desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we observe two women shaking hands at the reception desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 22.0], "base_se_norm": [0.13227975368499756, 0.24550771713256836], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.13227975368499756, 0.24550771713256836], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003846.pt"}
{"idx": 3848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-8qvGAWiPQY.mp4", "duration": 35.32695833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when a person is using a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a person is using a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [0.07566583156585693, 0.18889379501342773], "gt_se_raw": [19.0, 20.0], "gt_se_norm": [0.07566583156585693, 0.13227975368499756], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003848.pt"}
{"idx": 3850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RxAtuMu_ph4.mp4", "duration": 128.6285, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where cars are driving on a dirt road in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where cars are driving on a dirt road in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.5179917216300964, -0.4713457226753235], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.5024430751800537, -0.4713457226753235], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003850.pt"}
{"idx": 3852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RxAtuMu_ph4.mp4", "duration": 128.6285, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with a beard put on the red helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with a beard put on the red helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 93.0], "base_se_norm": [0.41492760181427, 0.44602489471435547], "gt_se_raw": [91.0, 94.0], "gt_se_norm": [0.41492760181427, 0.4615734815597534], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003852.pt"}
{"idx": 3854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-UggNH1YeRY.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman in red adds a large piece of beef to her shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman in red adds a large piece of beef to her shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 68.0], "base_se_norm": [-0.3471040725708008, -0.2839205861091614], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [-0.32604289054870605, -0.29445117712020874], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003854.pt"}
{"idx": 3856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-UggNH1YeRY.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman in red eat mayonnaise straight from the jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman in red eat mayonnaise straight from the jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 118.0], "base_se_norm": [0.16889441013336182, 0.2426084280014038], "gt_se_raw": [112.0, 117.0], "gt_se_norm": [0.17942488193511963, 0.23207783699035645], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003856.pt"}
{"idx": 3858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-UggNH1YeRY.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman in red pulls the timer off the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman in red pulls the timer off the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 169.0], "base_se_norm": [0.7480762004852295, 0.7796679735183716], "gt_se_raw": [167.0, 169.0], "gt_se_norm": [0.7586067914962769, 0.7796679735183716], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003858.pt"}
{"idx": 3860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6CfZwFzapCo.mp4", "duration": 113.19641666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the customer apologize after realizing the barista is deaf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the customer apologize after realizing the barista is deaf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [-0.2579270601272583, -0.22259026765823364], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [-0.2579270601272583, -0.22259026765823364], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003860.pt"}
{"idx": 3862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6CfZwFzapCo.mp4", "duration": 113.19641666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'DO THIS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'DO THIS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.469947874546051, -0.43461108207702637], "gt_se_raw": [31.0, 32.0], "gt_se_norm": [-0.4522795081138611, -0.43461108207702637], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003862.pt"}
{"idx": 3864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BlJaw7xc1bw.mp4", "duration": 190.2902902902903, "prompt_text": "Please find the visual event described by the sentence 'When is a person peeling a banana?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person peeling a banana?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.8108153343200684, -0.7582640647888184], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.8003051280975342, -0.7582640647888184], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003864.pt"}
{"idx": 3866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BlJaw7xc1bw.mp4", "duration": 190.2902902902903, "prompt_text": "Please find the visual event described by the sentence 'Show me when sugar is added to the mashed banana.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when sugar is added to the mashed banana.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 60.0], "base_se_norm": [-0.4744870662689209, -0.3693845272064209], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.4744870662689209, -0.45346659421920776], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003866.pt"}
{"idx": 3868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BlJaw7xc1bw.mp4", "duration": 190.2902902902903, "prompt_text": "Please find the visual event described by the sentence 'When is milk being poured into the glass bottle containing the banana mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is milk being poured into the glass bottle containing the banana mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 156.0], "base_se_norm": [0.5134772062301636, 0.6396002769470215], "gt_se_raw": [145.0, 155.0], "gt_se_norm": [0.5239874124526978, 0.6290900707244873], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003868.pt"}
{"idx": 3870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XwKm4RILZa4.mp4", "duration": 60.8608, "prompt_text": "Please find the visual event described by the sentence 'When is the signup page for a YouDescribe account shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the signup page for a YouDescribe account shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.507071852684021, -0.375624418258667], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.507071852684021, -0.375624418258667], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003870.pt"}
{"idx": 3872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XwKm4RILZa4.mp4", "duration": 60.8608, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the funding source for the project is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the funding source for the project is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 59.0], "base_se_norm": [0.47878432273864746, 0.9388505220413208], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [0.610231876373291, 0.7745411396026611], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003872.pt"}
{"idx": 3874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g4woIrtHTOc.mp4", "duration": 149.16, "prompt_text": "Please find the visual event described by the sentence 'When does the woman introduce herself as Laura?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman introduce herself as Laura?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.8659157752990723], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9865915775299072, -0.9597747325897217], "start_token_span": [69, 72], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003874.pt"}
{"idx": 3876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g4woIrtHTOc.mp4", "duration": 149.16, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman stand outside the ABS building of the University of Sydney?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman stand outside the ABS building of the University of Sydney?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 53.0], "base_se_norm": [-0.4100295305252075, -0.28935372829437256], "gt_se_raw": [44.0, 53.0], "gt_se_norm": [-0.4100295305252075, -0.28935372829437256], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003876.pt"}
{"idx": 3878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g4woIrtHTOc.mp4", "duration": 149.16, "prompt_text": "Please find the visual event described by the sentence 'When are the special chairs for wheelchair users shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the special chairs for wheelchair users shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 96.0], "base_se_norm": [0.032448410987854004, 0.28720831871032715], "gt_se_raw": [78.0, 91.0], "gt_se_norm": [0.04585671424865723, 0.22016620635986328], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003878.pt"}
{"idx": 3880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QoWskNBiHdE.mp4", "duration": 152.31883333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the term 'SYMBOLIC ANNIHILATION' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the term 'SYMBOLIC ANNIHILATION' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 78.0], "base_se_norm": [-0.04148423671722412, 0.024167537689208984], "gt_se_raw": [74.0, 76.0], "gt_se_norm": [-0.028353869915008545, -0.002093195915222168], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003880.pt"}
{"idx": 3882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QoWskNBiHdE.mp4", "duration": 152.31883333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the faces of the two friends covered by question marks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the faces of the two friends covered by question marks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 133.0], "base_se_norm": [0.6150333881378174, 0.7463369369506836], "gt_se_raw": [123.0, 132.0], "gt_se_norm": [0.6150333881378174, 0.7332065105438232], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003882.pt"}
{"idx": 3884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLackFf0Zjw.mp4", "duration": 79.78803333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the knight in black armor being handed a red lightsaber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the knight in black armor being handed a red lightsaber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.5738708972930908, -0.4986717104911804], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.548804521560669, -0.4986717104911804], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003884.pt"}
{"idx": 3886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLackFf0Zjw.mp4", "duration": 79.78803333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the knight in silver armor fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the knight in silver armor fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.19787466526031494, -0.14774185419082642], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.19787466526031494, -0.14774185419082642], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003886.pt"}
{"idx": 3888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLackFf0Zjw.mp4", "duration": 79.78803333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the king shouts in anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the king shouts in anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 63.0], "base_se_norm": [0.5039849281311035, 0.5791841745376587], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [0.5791841745376587, 0.629317045211792], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003888.pt"}
{"idx": 3890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpeXE9XCU3o.mp4", "duration": 54.541666666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man is taking a picture of a cityscape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man is taking a picture of a cityscape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.889992356300354, -0.7433155179023743], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.8533231616020203, -0.7433155179023743], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003890.pt"}
{"idx": 3892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpeXE9XCU3o.mp4", "duration": 54.541666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point is an image of an alien toy displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an image of an alien toy displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.13674557209014893, 0.24675321578979492], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [0.13674557209014893, 0.24675321578979492], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003892.pt"}
{"idx": 3894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZYYCSEV-i1Y.mp4", "duration": 151.15099999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in a purple coat walk on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in a purple coat walk on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.6956685781478882, -0.6295095682144165], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.6956685781478882, -0.642741322517395], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003894.pt"}
{"idx": 3896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZYYCSEV-i1Y.mp4", "duration": 151.15099999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two women are fighting for the boots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two women are fighting for the boots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 78.0], "base_se_norm": [-0.06054210662841797, 0.03208053112030029], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [0.005616903305053711, 0.03208053112030029], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003896.pt"}
{"idx": 3898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZYYCSEV-i1Y.mp4", "duration": 151.15099999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman tries to melt the ice with two hair dryers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman tries to melt the ice with two hair dryers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 130.0], "base_se_norm": [0.6936705112457275, 0.7201341390609741], "gt_se_raw": [128.0, 130.0], "gt_se_norm": [0.6936705112457275, 0.7201341390609741], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003898.pt"}
{"idx": 3900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/noGcxr9xiGE.mp4", "duration": 56.99026666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is the boy actively tying his shoelaces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is the boy actively tying his shoelaces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 39.0], "base_se_norm": [-0.40340685844421387, 0.36865484714508057], "gt_se_raw": [16.0, 38.0], "gt_se_norm": [-0.4385005831718445, 0.3335610628128052], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003900.pt"}
{"idx": 3902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/noGcxr9xiGE.mp4", "duration": 56.99026666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a close-up shows the successfully tied shoelaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a close-up shows the successfully tied shoelaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [0.5090296268463135, 0.6494045257568359], "gt_se_raw": [42.0, 47.0], "gt_se_norm": [0.47393596172332764, 0.6494045257568359], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003902.pt"}
{"idx": 3904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZAeKfGpd2M.mp4", "duration": 230.60537499999998, "prompt_text": "Please find the visual event described by the sentence 'When is the man stretching the green ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man stretching the green ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 43.0], "base_se_norm": [-0.9045989513397217, -0.6270685195922852], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.9045989513397217, -0.8785804510116577], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003904.pt"}
{"idx": 3906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZAeKfGpd2M.mp4", "duration": 230.60537499999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the man pounding the ice cream with a large wooden pestle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man pounding the ice cream with a large wooden pestle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 121.0], "base_se_norm": [-0.08068060874938965, 0.049411773681640625], "gt_se_raw": [113.0, 121.0], "gt_se_norm": [-0.019970834255218506, 0.049411773681640625], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003906.pt"}
{"idx": 3908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZAeKfGpd2M.mp4", "duration": 230.60537499999998, "prompt_text": "Please find the visual event described by the sentence 'When are the ingredients poured into a pot of milk and whisked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the ingredients poured into a pot of milk and whisked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 92.0], "base_se_norm": [-0.2454642653465271, -0.20210015773773193], "gt_se_raw": [87.0, 92.0], "gt_se_norm": [-0.2454642653465271, -0.20210015773773193], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003908.pt"}
{"idx": 3910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4swYF9Szvs.mp4", "duration": 173.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man standing in his office, looking confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man standing in his office, looking confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 42.0], "base_se_norm": [-0.6318312883377075, -0.5167784690856934], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.6318312883377075, -0.5282837748527527], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003910.pt"}
{"idx": 3912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4swYF9Szvs.mp4", "duration": 173.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where a woman is relaxing on a beach chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where a woman is relaxing on a beach chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 77.0], "base_se_norm": [-0.2176414132118225, -0.11409395933151245], "gt_se_raw": [68.0, 77.0], "gt_se_norm": [-0.2176414132118225, -0.11409395933151245], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003912.pt"}
{"idx": 3914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jF3I7VpfCEU.mp4", "duration": 174.96645555555554, "prompt_text": "Please find the visual event described by the sentence 'When can we see the answers 'Exercising' and 'Stretching out like RICHARD SIMMONS'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the answers 'Exercising' and 'Stretching out like RICHARD SIMMONS'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.8171077966690063, -0.7713847756385803], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.805677056312561, -0.7713847756385803], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003914.pt"}
{"idx": 3916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jF3I7VpfCEU.mp4", "duration": 174.96645555555554, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the answers 'Yoga' and 'Juggling!' are revealed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the answers 'Yoga' and 'Juggling!' are revealed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 81.0], "base_se_norm": [-0.14269286394119263, -0.07410824298858643], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [-0.13126206398010254, -0.07410824298858643], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003916.pt"}
{"idx": 3918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jF3I7VpfCEU.mp4", "duration": 174.96645555555554, "prompt_text": "Please find the visual event described by the sentence 'Show the part where one man turns his head to check the other's eye color.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 153.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where one man turns his head to check the other's eye color.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 153.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [153.0, 163.0], "base_se_norm": [0.7489066123962402, 0.8632142543792725], "gt_se_raw": [159.0, 162.0], "gt_se_norm": [0.8174911737442017, 0.8517835140228271], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003918.pt"}
{"idx": 3920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qI34nBlJxP8.mp4", "duration": 104.06229166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Woodstock and his friends decorate the doghouse with a star and lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Woodstock and his friends decorate the doghouse with a star and lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.6156148910522461, -0.4618607759475708], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.5963956117630005, -0.5387378334999084], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003920.pt"}
{"idx": 3922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qI34nBlJxP8.mp4", "duration": 104.06229166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the little yellow birds carry the string of lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the little yellow birds carry the string of lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.7693688869476318, -0.6540533304214478], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.7309303879737854, -0.692491888999939], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003922.pt"}
{"idx": 3924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y4yZelVCE3o.mp4", "duration": 180.52, "prompt_text": "Please find the visual event described by the sentence 'When can we see the family hiding under the dining table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the family hiding under the dining table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 29.0], "base_se_norm": [-0.8670507669448853, -0.6787059307098389], "gt_se_raw": [13.0, 22.0], "gt_se_norm": [-0.85597163438797, -0.7562596797943115], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003924.pt"}
{"idx": 3926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y4yZelVCE3o.mp4", "duration": 180.52, "prompt_text": "Please find the visual event described by the sentence 'At what point does the black car make a jump over a large gap in the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the black car make a jump over a large gap in the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 104.0], "base_se_norm": [0.11898958683013916, 0.15222692489624023], "gt_se_raw": [102.0, 107.0], "gt_se_norm": [0.13006865978240967, 0.18546414375305176], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003926.pt"}
{"idx": 3928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y4yZelVCE3o.mp4", "duration": 180.52, "prompt_text": "Please find the visual event described by the sentence 'When does the glass skyscraper begin to collapse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 165.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the glass skyscraper begin to collapse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 165.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 165.0], "base_se_norm": [0.7837358713150024, 0.828052282333374], "gt_se_raw": [161.0, 166.0], "gt_se_norm": [0.7837358713150024, 0.8391313552856445], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003928.pt"}
{"idx": 3930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D0a0aNqTehM.mp4", "duration": 360.72703333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the caterpillar fails to wear the cocoon and falls to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the caterpillar fails to wear the cocoon and falls to the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 156.0], "base_se_norm": [-0.15725749731063843, -0.13508003950119019], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.7837700247764587, -0.7615925669670105], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003930.pt"}
{"idx": 3932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D0a0aNqTehM.mp4", "duration": 360.72703333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the two insects swing to the cocoon on a stone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 211.0 - 245.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the two insects swing to the cocoon on a stone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 211.0 - 245.0 seconds.<|im_end|>\n", "base_se_raw": [211.0, 245.0], "base_se_norm": [0.1698596477508545, 0.358367919921875], "gt_se_raw": [223.0, 242.0], "gt_se_norm": [0.23639202117919922, 0.34173476696014404], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003932.pt"}
{"idx": 3934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pjJEXkbeL-o.mp4", "duration": 278.76, "prompt_text": "Please find the visual event described by the sentence 'When is the man with glasses and a cigar laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man with glasses and a cigar laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.6484431028366089, -0.6269192695617676], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.6484431028366089, -0.6340938806533813], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003934.pt"}
{"idx": 3936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pjJEXkbeL-o.mp4", "duration": 278.76, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man in the uniform is looking at his reflection in the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man in the uniform is looking at his reflection in the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 138.0], "base_se_norm": [-0.26101309061050415, -0.009901046752929688], "gt_se_raw": [117.0, 128.0], "gt_se_norm": [-0.1605682373046875, -0.08164733648300171], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003936.pt"}
{"idx": 3938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pjJEXkbeL-o.mp4", "duration": 278.76, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man in the uniform wave a white handkerchief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 247.0 - 252.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man in the uniform wave a white handkerchief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 247.0 - 252.0 seconds.<|im_end|>\n", "base_se_raw": [247.0, 252.0], "base_se_norm": [0.7721337080001831, 0.8080068826675415], "gt_se_raw": [250.0, 252.0], "gt_se_norm": [0.7936575412750244, 0.8080068826675415], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003938.pt"}
{"idx": 3940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fNINO2RnYz0.mp4", "duration": 14.014000000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me the loading screen with the text 'Yume Lion OS 4.2'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the loading screen with the text 'Yume Lion OS 4.2'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [0.2844299077987671, 0.8552875518798828], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [0.141715407371521, 0.8552875518798828], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003940.pt"}
{"idx": 3942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fNINO2RnYz0.mp4", "duration": 14.014000000000001, "prompt_text": "Please find the visual event described by the sentence 'At what time does the 'loading complete' chime play?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the 'loading complete' chime play?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 13.0], "base_se_norm": [0.7125731706619263, 0.8552875518798828], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [0.8552875518798828, 0.9980020523071289], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003942.pt"}
{"idx": 3944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/USGLDwaLnaI.mp4", "duration": 88.75533333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man pushing a cart along the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man pushing a cart along the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 21.0], "base_se_norm": [-0.9323984384536743, -0.5267890095710754], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.7295937538146973, -0.6169244050979614], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003944.pt"}
{"idx": 3946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/USGLDwaLnaI.mp4", "duration": 88.75533333333334, "prompt_text": "Please find the visual event described by the sentence 'At what time are there a group of people gathered around a cart with a red umbrella?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are there a group of people gathered around a cart with a red umbrella?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [-0.25638270378112793, -0.09864568710327148], "gt_se_raw": [34.0, 40.0], "gt_se_norm": [-0.23384886980056763, -0.09864568710327148], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003946.pt"}
{"idx": 3948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/USGLDwaLnaI.mp4", "duration": 88.75533333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman with glasses is smiling while holding a blue shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman with glasses is smiling while holding a blue shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 61.0], "base_se_norm": [0.3069636821746826, 0.3745652437210083], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [0.3069636821746826, 0.3745652437210083], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003948.pt"}
{"idx": 3950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GYDaVvjI5kY.mp4", "duration": 202.06853333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the blonde woman screams because there's a lizard on her water bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the blonde woman screams because there's a lizard on her water bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [-0.2576775550842285, -0.20818942785263062], "gt_se_raw": [76.0, 80.0], "gt_se_norm": [-0.2477799654006958, -0.20818942785263062], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003950.pt"}
{"idx": 3952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GYDaVvjI5kY.mp4", "duration": 202.06853333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman in black threaten the two girls, telling them she will make their lives miserable?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman in black threaten the two girls, telling them she will make their lives miserable?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 148.0], "base_se_norm": [0.30648744106292725, 0.464849591255188], "gt_se_raw": [133.0, 147.0], "gt_se_norm": [0.3163851499557495, 0.4549520015716553], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003952.pt"}
{"idx": 3954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ur83i6_BjbE.mp4", "duration": 143.26812222222222, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in the purple shirt score a basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in the purple shirt score a basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.790602445602417, -0.7487229108810425], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.7766425609588623, -0.7487229108810425], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003954.pt"}
{"idx": 3956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ur83i6_BjbE.mp4", "duration": 143.26812222222222, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two people are sitting on the bleachers at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two people are sitting on the bleachers at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 76.0], "base_se_norm": [-0.008851408958435059, 0.06094777584075928], "gt_se_raw": [71.0, 76.0], "gt_se_norm": [-0.008851408958435059, 0.06094777584075928], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003956.pt"}
{"idx": 3958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ur83i6_BjbE.mp4", "duration": 143.26812222222222, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the couple is playing strip basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the couple is playing strip basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 64.0], "base_se_norm": [-0.21824902296066284, -0.106570303440094], "gt_se_raw": [55.0, 62.0], "gt_se_norm": [-0.23220884799957275, -0.13448995351791382], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003958.pt"}
{"idx": 3960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/83Wu5xmstn4.mp4", "duration": 132.88275, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a person makes a slam dunk in basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a person makes a slam dunk in basketball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.5183724164962769, -0.48827069997787476], "gt_se_raw": [33.0, 34.0], "gt_se_norm": [-0.5033215284347534, -0.48827069997787476], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003960.pt"}
{"idx": 3962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/83Wu5xmstn4.mp4", "duration": 132.88275, "prompt_text": "Please find the visual event described by the sentence 'What is the time range showing a lion mascot posing with a family?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range showing a lion mascot posing with a family?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 74.0], "base_se_norm": [0.08366203308105469, 0.11376380920410156], "gt_se_raw": [72.0, 73.0], "gt_se_norm": [0.08366203308105469, 0.09871292114257812], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003962.pt"}
{"idx": 3964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xoPJu25ZHxM.mp4", "duration": 273.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the 'Princess Academy' paper burned with a cigarette holder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'Princess Academy' paper burned with a cigarette holder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.9050517082214355, -0.8758369088172913], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.9050517082214355, -0.8758369088172913], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003964.pt"}
{"idx": 3966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xoPJu25ZHxM.mp4", "duration": 273.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man wearing a top hat with a skull and crossbones appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man wearing a top hat with a skull and crossbones appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 162.0], "base_se_norm": [0.11016428470611572, 0.18320143222808838], "gt_se_raw": [152.0, 156.0], "gt_se_norm": [0.11016428470611572, 0.13937914371490479], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003966.pt"}
{"idx": 3968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xoPJu25ZHxM.mp4", "duration": 273.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when all five villains are laughing heartily together at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 247.0 - 251.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when all five villains are laughing heartily together at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 247.0 - 251.0 seconds.<|im_end|>\n", "base_se_raw": [247.0, 251.0], "base_se_norm": [0.8040169477462769, 0.8332318067550659], "gt_se_raw": [247.0, 251.0], "gt_se_norm": [0.8040169477462769, 0.8332318067550659], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003968.pt"}
{"idx": 3970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBiE0ltuFko.mp4", "duration": 130.1, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the girl is waving at her shadow on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the girl is waving at her shadow on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 102.0], "base_se_norm": [0.4296693801879883, 0.5680245161056519], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8616448640823364, -0.8001537322998047], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003970.pt"}
{"idx": 3972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBiE0ltuFko.mp4", "duration": 130.1, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl put her hand right in front of the camera lens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl put her hand right in front of the camera lens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 68.0], "base_se_norm": [-0.01614147424697876, 0.045349717140197754], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.0007686614990234375, 0.045349717140197754], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003972.pt"}
{"idx": 3974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uQp0Eihw2WA.mp4", "duration": 60.026633333333336, "prompt_text": "Please find the visual event described by the sentence 'When does the girl say 'I love you mama'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl say 'I love you mama'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.5335403680801392, -0.46690332889556885], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.5668589472770691, -0.500221848487854], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003974.pt"}
{"idx": 3976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uQp0Eihw2WA.mp4", "duration": 60.026633333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl touch her head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl touch her head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [0.09951198101043701, 0.19946753978729248], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [0.09951198101043701, 0.19946753978729248], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003976.pt"}
{"idx": 3978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uQp0Eihw2WA.mp4", "duration": 60.026633333333336, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the girl crosses her arms in front of her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the girl crosses her arms in front of her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [0.23278605937957764, 0.3660602569580078], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [0.23278605937957764, 0.33274173736572266], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003978.pt"}
{"idx": 3980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0Ue-JMqh0CE.mp4", "duration": 17.684333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the person in the blue sweatshirt is playing the drums.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the person in the blue sweatshirt is playing the drums.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.6964167356491089], "gt_se_raw": [0.0, 16.0], "gt_se_norm": [-1.0, 0.8095111846923828], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003980.pt"}
{"idx": 3982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0Ue-JMqh0CE.mp4", "duration": 17.684333333333335, "prompt_text": "Please find the visual event described by the sentence 'When do the drum pads light up with a blue/cyan color as they are being played?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the drum pads light up with a blue/cyan color as they are being played?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, 0.8095111846923828], "gt_se_raw": [0.0, 16.0], "gt_se_norm": [-1.0, 0.8095111846923828], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003982.pt"}
{"idx": 3984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zkCoaPOsw5U.mp4", "duration": 197.0, "prompt_text": "Please find the visual event described by the sentence 'When is the woman reading an e-book in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman reading an e-book in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 27.0], "base_se_norm": [-0.796954333782196, -0.7258883118629456], "gt_se_raw": [20.0, 27.0], "gt_se_norm": [-0.796954333782196, -0.7258883118629456], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003984.pt"}
{"idx": 3986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zkCoaPOsw5U.mp4", "duration": 197.0, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the woman is jogging along the lakeside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the woman is jogging along the lakeside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 107.0], "base_se_norm": [-0.2588832378387451, 0.08629441261291504], "gt_se_raw": [86.0, 98.0], "gt_se_norm": [-0.12690353393554688, -0.005076169967651367], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003986.pt"}
{"idx": 3988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zkCoaPOsw5U.mp4", "duration": 197.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman shown pouring the blended watermelon juice into glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman shown pouring the blended watermelon juice into glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 155.0], "base_se_norm": [0.5431472063064575, 0.5736041069030762], "gt_se_raw": [152.0, 155.0], "gt_se_norm": [0.5431472063064575, 0.5736041069030762], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003988.pt"}
{"idx": 3990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VIglbPLTzVs.mp4", "duration": 23.289933333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the exact moment the person flips the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the exact moment the person flips the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.4847559332847595, -0.05538588762283325], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.4847559332847595, -0.3130079507827759], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003990.pt"}
{"idx": 3992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VIglbPLTzVs.mp4", "duration": 23.289933333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the person reading the text on the left page, after turning to the page with the illustration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person reading the text on the left page, after turning to the page with the illustration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [0.11636209487915039, 0.7174801826477051], "gt_se_raw": [9.0, 19.0], "gt_se_norm": [-0.22713392972946167, 0.6316061019897461], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003992.pt"}
{"idx": 3994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JEDrAOt0er8.mp4", "duration": 94.55279166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy with the green shirt become angry?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy with the green shirt become angry?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 17.0], "base_se_norm": [-0.8096301555633545, -0.6404125690460205], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.8307823538780212, -0.767325758934021], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003994.pt"}
{"idx": 3996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JEDrAOt0er8.mp4", "duration": 94.55279166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the peacock-shaped gramophone appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the peacock-shaped gramophone appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 46.0], "base_se_norm": [-0.17506402730941772, -0.02699863910675049], "gt_se_raw": [39.0, 46.0], "gt_se_norm": [-0.17506402730941772, -0.02699863910675049], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003996.pt"}
{"idx": 3998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JEDrAOt0er8.mp4", "duration": 94.55279166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the explosion happen on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the explosion happen on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 78.0], "base_se_norm": [0.607567548751831, 0.6498719453811646], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.5864152908325195, 0.6498719453811646], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003998.pt"}
{"idx": 4000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ho8L7wzkeG0.mp4", "duration": 92.29220000000001, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is talking to the camera while sitting in a restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is talking to the camera while sitting in a restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.34989088773727417, -0.26320964097976685], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.34989088773727417, -0.26320964097976685], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004000.pt"}
{"idx": 4002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ho8L7wzkeG0.mp4", "duration": 92.29220000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the red beanie hug another woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the red beanie hug another woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 48.0], "base_se_norm": [0.01850426197052002, 0.04017460346221924], "gt_se_raw": [47.0, 48.0], "gt_se_norm": [0.01850426197052002, 0.04017460346221924], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004002.pt"}
{"idx": 4004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3h6KMumLAvI.mp4", "duration": 15.849166666666667, "prompt_text": "Please find the visual event described by the sentence 'When is a card being slid from the center of the board to the right side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a card being slid from the center of the board to the right side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 12.0], "base_se_norm": [-0.6214312314987183, 0.5142751932144165], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.24286240339279175, 0.00951683521270752], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004004.pt"}
{"idx": 4006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3h6KMumLAvI.mp4", "duration": 15.849166666666667, "prompt_text": "Please find the visual event described by the sentence 'When can the words 'by yourself' be heard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can the words 'by yourself' be heard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.261896014213562, 0.5142751932144165], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.00951683521270752, 0.13570642471313477], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004006.pt"}
{"idx": 4008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3h6KMumLAvI.mp4", "duration": 15.849166666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker describe an activity to do with Dustin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker describe an activity to do with Dustin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.8738104104995728, 0.261896014213562], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [0.261896014213562, 0.5142751932144165], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004008.pt"}
{"idx": 4010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gsYL4PC0hyk.mp4", "duration": 253.16958333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where one stick figure is pushing a dark monster-like creature into a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where one stick figure is pushing a dark monster-like creature into a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 94.0], "base_se_norm": [-0.2732144594192505, -0.2574147582054138], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [-0.2732144594192505, -0.24951493740081787], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004010.pt"}
{"idx": 4012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gsYL4PC0hyk.mp4", "duration": 253.16958333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display an hourglass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display an hourglass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.8420031070709229, -0.8183035850524902], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.8420031070709229, -0.8183035850524902], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004012.pt"}
{"idx": 4014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMmIMQuqVa0.mp4", "duration": 136.87006666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a boy in a red shirt playing with bells?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a boy in a red shirt playing with bells?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.9561628103256226, -0.853875994682312], "gt_se_raw": [5.0, 11.0], "gt_se_norm": [-0.926937997341156, -0.8392636179924011], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004014.pt"}
{"idx": 4016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMmIMQuqVa0.mp4", "duration": 136.87006666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a boy in a wheelchair is given a cloth to play with.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a boy in a wheelchair is given a cloth to play with.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 45.0], "base_se_norm": [-0.5470156669616699, -0.3424420952796936], "gt_se_raw": [32.0, 46.0], "gt_se_norm": [-0.5324032306671143, -0.32782965898513794], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004016.pt"}
{"idx": 4018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMmIMQuqVa0.mp4", "duration": 136.87006666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the little girl in the pink shirt hugs the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the little girl in the pink shirt hugs the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 103.0], "base_se_norm": [0.35895299911499023, 0.5050770044326782], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [0.7242629528045654, 0.7681001424789429], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004018.pt"}
{"idx": 4020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xdr_rWE8qOQ.mp4", "duration": 195.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the duck watches a video on a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the duck watches a video on a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [-0.7744361162185669, -0.7539302706718445], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.7744361162185669, -0.7539302706718445], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004020.pt"}
{"idx": 4022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xdr_rWE8qOQ.mp4", "duration": 195.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the duck standing in a yellow tub?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the duck standing in a yellow tub?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.7539302706718445, -0.723171591758728], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.7436773777008057, -0.723171591758728], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004022.pt"}
{"idx": 4024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CMYlBq83gu4.mp4", "duration": 155.155, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the man is covering one eye for an eye test.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the man is covering one eye for an eye test.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.5617285966873169, -0.5101672410964966], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.5617285966873169, -0.5101672410964966], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004024.pt"}
{"idx": 4026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CMYlBq83gu4.mp4", "duration": 155.155, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man using a handheld electronic magnifier to read a document?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man using a handheld electronic magnifier to read a document?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [-0.32970255613327026, -0.2652508616447449], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.3039218783378601, -0.2523605227470398], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004026.pt"}
{"idx": 4028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CMYlBq83gu4.mp4", "duration": 155.155, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the woman and the man are looking at a spiral-bound booklet together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the woman and the man are looking at a spiral-bound booklet together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.3683735728263855, -0.3168122172355652], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.3683735728263855, -0.32970255613327026], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004028.pt"}
{"idx": 4030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghg2AP3i5TI.mp4", "duration": 34.26756666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a woman in a light blue shirt writes on a yellow sticky note.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a woman in a light blue shirt writes on a yellow sticky note.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.45910561084747314, 0.6341983079910278], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [0.5174698829650879, 0.5758341550827026], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004030.pt"}
{"idx": 4032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghg2AP3i5TI.mp4", "duration": 34.26756666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a man in a light blue button-down shirt holds up a large sheet of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a man in a light blue button-down shirt holds up a large sheet of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.7081788778305054, -0.299629271030426], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.7081788778305054, -0.6498146057128906], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004032.pt"}
{"idx": 4034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bhA2zrVBikA.mp4", "duration": 128.9, "prompt_text": "Please find the visual event described by the sentence 'When does the woman on the right pick up a stack of papers and place it in front of her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman on the right pick up a stack of papers and place it in front of her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [0.008533835411071777, 0.11714518070220947], "gt_se_raw": [82.0, 86.0], "gt_se_norm": [0.2723041772842407, 0.3343677520751953], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004034.pt"}
{"idx": 4036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bhA2zrVBikA.mp4", "duration": 128.9, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman on the right rests her head on the table, appearing tired.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman on the right rests her head on the table, appearing tired.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 119.0], "base_se_norm": [0.7688131332397461, 0.8463926315307617], "gt_se_raw": [114.0, 119.0], "gt_se_norm": [0.7688131332397461, 0.8463926315307617], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004036.pt"}
{"idx": 4038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bhA2zrVBikA.mp4", "duration": 128.9, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman on the right stops working and leans back in her chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman on the right stops working and leans back in her chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 46.0], "base_se_norm": [-0.44142740964889526, -0.286268413066864], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [-0.4569433331489563, -0.33281612396240234], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004038.pt"}
{"idx": 4040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t-WR-8-HEx4.mp4", "duration": 31.197833333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man touch the green blocks on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man touch the green blocks on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9358929991722107, -0.5512509346008301], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.8717859983444214, -0.6153578758239746], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004040.pt"}
{"idx": 4042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t-WR-8-HEx4.mp4", "duration": 31.197833333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the man use his white cane to locate the edge of the door frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man use his white cane to locate the edge of the door frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 29.0], "base_se_norm": [0.47446131706237793, 0.8591034412384033], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [0.6667823791503906, 0.7949963808059692], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004042.pt"}
{"idx": 4044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nuavO7Of92M.mp4", "duration": 142.97616666666664, "prompt_text": "Please find the visual event described by the sentence 'When do the two speakers introduce themselves by name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two speakers introduce themselves by name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.9020816087722778, -0.8181514739990234], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9020816087722778, -0.8741048574447632], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004044.pt"}
{"idx": 4046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nuavO7Of92M.mp4", "duration": 142.97616666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point is a close-up shot of a clock shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a close-up shot of a clock shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.45445454120635986, -0.3985011577606201], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.44046616554260254, -0.41248947381973267], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004046.pt"}
{"idx": 4048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nuavO7Of92M.mp4", "duration": 142.97616666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the man mention the admission requirement of 20 hours of experience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man mention the admission requirement of 20 hours of experience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 101.0], "base_se_norm": [0.30091607570648193, 0.4128228425979614], "gt_se_raw": [98.0, 100.0], "gt_se_norm": [0.370857834815979, 0.39883458614349365], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004048.pt"}
{"idx": 4050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksdOh7MbRgk.mp4", "duration": 178.91206666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when two boys are sitting on the floor playing with colorful building blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when two boys are sitting on the floor playing with colorful building blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.4522448778152466, -0.40753018856048584], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.4522448778152466, -0.418708860874176], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004050.pt"}
{"idx": 4052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksdOh7MbRgk.mp4", "duration": 178.91206666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman and a boy taking a tray of cookies out of the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman and a boy taking a tray of cookies out of the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 121.0], "base_se_norm": [0.3190838098526001, 0.3526197671890259], "gt_se_raw": [117.0, 121.0], "gt_se_norm": [0.30790507793426514, 0.3526197671890259], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004052.pt"}
{"idx": 4054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFjp1MW6Lzc.mp4", "duration": 119.96, "prompt_text": "Please find the visual event described by the sentence 'Show me the man cycling during the sunrise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man cycling during the sunrise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 21.0], "base_se_norm": [-0.9499833583831787, -0.6498832702636719], "gt_se_raw": [3.0, 10.0], "gt_se_norm": [-0.9499833583831787, -0.8332777619361877], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004054.pt"}
{"idx": 4056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFjp1MW6Lzc.mp4", "duration": 119.96, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a person's bare feet are shown on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a person's bare feet are shown on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 69.0], "base_se_norm": [0.01700568199157715, 0.15038347244262695], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.03367793560028076, 0.11703896522521973], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004056.pt"}
{"idx": 4058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFjp1MW6Lzc.mp4", "duration": 119.96, "prompt_text": "Please find the visual event described by the sentence 'When does the boy on the bicycle pass by a red car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy on the bicycle pass by a red car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.2664221525192261, -0.19973325729370117], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.2830943465232849, -0.21640545129776], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004058.pt"}
{"idx": 4060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CVtTOpgSyY.mp4", "duration": 30.833333333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a boat is sailing on the sea under the full moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a boat is sailing on the sea under the full moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 12.0], "base_se_norm": [-0.545945942401886, -0.22162163257598877], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.545945942401886, -0.22162163257598877], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004060.pt"}
{"idx": 4062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CVtTOpgSyY.mp4", "duration": 30.833333333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the boy is running across the field of stars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the boy is running across the field of stars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [0.16756749153137207, 0.42702698707580566], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [0.16756749153137207, 0.42702698707580566], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004062.pt"}
{"idx": 4064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRi80V8ire8.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman touching an abstract painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman touching an abstract painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.8001998066902161], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.8001998066902161], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004064.pt"}
{"idx": 4066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRi80V8ire8.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman walking with her dog in an outdoor restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman walking with her dog in an outdoor restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.22577422857284546, -0.15084916353225708], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.25074928998947144, -0.15084916353225708], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004066.pt"}
{"idx": 4068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRi80V8ire8.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman applying the foam body wash to her arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman applying the foam body wash to her arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 66.0], "base_se_norm": [0.49850142002105713, 0.6483516693115234], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [0.5734264850616455, 0.6483516693115234], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004068.pt"}
{"idx": 4070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xWggTb45brM.mp4", "duration": 311.5, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man is dancing in the black and white hall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man is dancing in the black and white hall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 85.0], "base_se_norm": [-0.5955055952072144, -0.454253613948822], "gt_se_raw": [64.0, 77.0], "gt_se_norm": [-0.5890851020812988, -0.5056179761886597], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004070.pt"}
{"idx": 4072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xWggTb45brM.mp4", "duration": 311.5, "prompt_text": "Please find the visual event described by the sentence 'When does the man walk past the indoor pool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 184.0 - 197.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man walk past the indoor pool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 184.0 - 197.0 seconds.<|im_end|>\n", "base_se_raw": [184.0, 197.0], "base_se_norm": [0.18138039112091064, 0.2648475170135498], "gt_se_raw": [183.0, 188.0], "gt_se_norm": [0.17495989799499512, 0.20706260204315186], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004072.pt"}
{"idx": 4074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/81kd2SbB0qM.mp4", "duration": 31.5315, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the blue shirt guiding the boy in the red shirt through the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the blue shirt guiding the boy in the red shirt through the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 30.0], "base_se_norm": [-0.9365713596343994, 0.902859091758728], "gt_se_raw": [1.0, 28.0], "gt_se_norm": [-0.9365713596343994, 0.7760018110275269], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004074.pt"}
{"idx": 4076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/81kd2SbB0qM.mp4", "duration": 31.5315, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the wooden chest of drawers is clearly visible on the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the wooden chest of drawers is clearly visible on the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 30.0], "base_se_norm": [0.4588586091995239, 0.902859091758728], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.33200132846832275, 0.7760018110275269], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004076.pt"}
{"idx": 4078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/81kd2SbB0qM.mp4", "duration": 31.5315, "prompt_text": "Please find the visual event described by the sentence 'When are the two individuals in the initial part of the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two individuals in the initial part of the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9365713596343994, -0.3657136559486389], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9365713596343994, -0.4925709366798401], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004078.pt"}
{"idx": 4080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4dgJoFwlhE.mp4", "duration": 113.48003333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when two men are writing at a table and a woman is guiding them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when two men are writing at a table and a woman is guiding them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 96.0], "base_se_norm": [0.568557858467102, 0.6919275522232056], "gt_se_raw": [89.0, 96.0], "gt_se_norm": [0.568557858467102, 0.6919275522232056], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004080.pt"}
{"idx": 4082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4dgJoFwlhE.mp4", "duration": 113.48003333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a woman is sitting in front of a green screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a woman is sitting in front of a green screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 43.0], "base_se_norm": [-0.400775671005249, -0.2421574592590332], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [-0.38315141201019287, -0.2421574592590332], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004082.pt"}
{"idx": 4084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oM_M_d9OiHc.mp4", "duration": 75.3, "prompt_text": "Please find the visual event described by the sentence 'When can we see an aerial view of the coastline transformed by the red tide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see an aerial view of the coastline transformed by the red tide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.7609561681747437], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8937582969665527], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004084.pt"}
{"idx": 4086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oM_M_d9OiHc.mp4", "duration": 75.3, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of an elderly man expressing his surprise at the red water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of an elderly man expressing his surprise at the red water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.2563081383705139, -0.12350600957870483], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [-0.28286856412887573, -0.09694558382034302], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004086.pt"}
{"idx": 4088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oM_M_d9OiHc.mp4", "duration": 75.3, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where an expert collects a sample of the red water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where an expert collects a sample of the red water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 57.0], "base_se_norm": [0.3545815944671631, 0.513944149017334], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [0.3811420202255249, 0.48738372325897217], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004088.pt"}
{"idx": 4090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j4vo5y0eHnA.mp4", "duration": 81.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the children are all hiding under the inflated parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the children are all hiding under the inflated parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 73.0], "base_se_norm": [-0.7540983557701111, 0.7950818538665771], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.7295081615447998, -0.631147563457489], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004090.pt"}
{"idx": 4092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j4vo5y0eHnA.mp4", "duration": 81.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do the people outside shout 'Happy Birthday' to the children inside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the people outside shout 'Happy Birthday' to the children inside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.5081967115402222, -0.262295126914978], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [0.5, 0.5737704038619995], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004092.pt"}
{"idx": 4094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5N2BC6PX3Us.mp4", "duration": 141.8, "prompt_text": "Please find the visual event described by the sentence 'When can we see the audiogram for 'Normal Hearing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the audiogram for 'Normal Hearing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8448519110679626, -0.7743300199508667], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.8448519110679626, -0.7602256536483765], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004094.pt"}
{"idx": 4096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5N2BC6PX3Us.mp4", "duration": 141.8, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when Article 30 from the United Nations Convention is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when Article 30 from the United Nations Convention is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 65.0], "base_se_norm": [-0.4358251094818115, -0.08321583271026611], "gt_se_raw": [40.0, 66.0], "gt_se_norm": [-0.4358251094818115, -0.06911146640777588], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004096.pt"}
{"idx": 4098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5N2BC6PX3Us.mp4", "duration": 141.8, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman with pink hair signing on stage next to a shirtless singer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman with pink hair signing on stage next to a shirtless singer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 115.0], "base_se_norm": [0.5373765230178833, 0.6220028400421143], "gt_se_raw": [109.0, 116.0], "gt_se_norm": [0.5373765230178833, 0.6361072063446045], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004098.pt"}
{"idx": 4100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iWqNo5DnjtA.mp4", "duration": 107.34056666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the ginger is being smashed with the flat side of the knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the ginger is being smashed with the flat side of the knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 82.0], "base_se_norm": [0.15520167350769043, 0.5278472900390625], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [0.11793708801269531, 0.173833966255188], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004100.pt"}
{"idx": 4102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iWqNo5DnjtA.mp4", "duration": 107.34056666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man is mincing the crushed ginger with a rocking motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man is mincing the crushed ginger with a rocking motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 88.0], "base_se_norm": [0.30425989627838135, 0.6396410465240479], "gt_se_raw": [78.0, 82.0], "gt_se_norm": [0.4533182382583618, 0.5278472900390625], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004102.pt"}
{"idx": 4104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1P6zkiJ6EQA.mp4", "duration": 60.76067734401068, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman wiping a laptop with a disinfectant wipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman wiping a laptop with a disinfectant wipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 26.0], "base_se_norm": [-0.5391756296157837, -0.14418333768844604], "gt_se_raw": [17.0, 26.0], "gt_se_norm": [-0.4404275417327881, -0.14418333768844604], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004104.pt"}
{"idx": 4106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1P6zkiJ6EQA.mp4", "duration": 60.76067734401068, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a black shirt walk towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a black shirt walk towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [0.2508089542388916, 0.4153890609741211], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [0.2837250232696533, 0.4153890609741211], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004106.pt"}
{"idx": 4108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1P6zkiJ6EQA.mp4", "duration": 60.76067734401068, "prompt_text": "Please find the visual event described by the sentence 'When is the 'Circulation' sign shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'Circulation' sign shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 55.0], "base_se_norm": [0.7445493936538696, 0.8103814125061035], "gt_se_raw": [54.0, 55.0], "gt_se_norm": [0.7774653434753418, 0.8103814125061035], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004108.pt"}
{"idx": 4110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4sx42_C10zw.mp4", "duration": 105.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the sentence about going to college is analyzed into topic, claim, and specific points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the sentence about going to college is analyzed into topic, claim, and specific points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 50.0], "base_se_norm": [-0.48571425676345825, -0.0476190447807312], "gt_se_raw": [27.0, 50.0], "gt_se_norm": [-0.48571425676345825, -0.0476190447807312], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004110.pt"}
{"idx": 4112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4sx42_C10zw.mp4", "duration": 105.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a red heart is drawn on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a red heart is drawn on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 88.0], "base_se_norm": [0.6190476417541504, 0.6761904954910278], "gt_se_raw": [86.0, 88.0], "gt_se_norm": [0.6380952596664429, 0.6761904954910278], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004112.pt"}
{"idx": 4114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IPPs6cIFZqM.mp4", "duration": 491.62446666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the woman pour the drink from the bottle into the glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman pour the drink from the bottle into the glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.7681156992912292, -0.7477749586105347], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [-0.7681156992912292, -0.7518430948257446], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004114.pt"}
{"idx": 4116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IPPs6cIFZqM.mp4", "duration": 491.62446666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman is blending foundation all over her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman is blending foundation all over her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 185.0], "base_se_norm": [-0.3735055923461914, -0.24739307165145874], "gt_se_raw": [161.0, 181.0], "gt_se_norm": [-0.34502851963043213, -0.2636656165122986], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004116.pt"}
{"idx": 4118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IPPs6cIFZqM.mp4", "duration": 491.62446666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman apply mascara to her eyelashes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 307.0 - 337.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman apply mascara to her eyelashes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 307.0 - 337.0 seconds.<|im_end|>\n", "base_se_raw": [307.0, 337.0], "base_se_norm": [0.24892067909240723, 0.3709651231765747], "gt_se_raw": [359.0, 375.0], "gt_se_norm": [0.4604642391204834, 0.5255546569824219], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004118.pt"}
{"idx": 4120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q6rAllJAdWk.mp4", "duration": 176.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man receives his order from the employee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man receives his order from the employee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 26.0], "base_se_norm": [-0.9320755004882812, -0.705660343170166], "gt_se_raw": [7.0, 20.0], "gt_se_norm": [-0.9207547307014465, -0.7735849022865295], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004120.pt"}
{"idx": 4122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q6rAllJAdWk.mp4", "duration": 176.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the 'BADLANDSCHUGS' logo appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'BADLANDSCHUGS' logo appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [-0.649056613445282, -0.592452883720398], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.6377358436584473, -0.592452883720398], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004122.pt"}
{"idx": 4124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6yYd6Pq7Ic.mp4", "duration": 64.25, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the green suit happily spinning in the revolving door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the green suit happily spinning in the revolving door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [0.6809338331222534, 0.929961085319519], "gt_se_raw": [52.0, 60.0], "gt_se_norm": [0.618677020072937, 0.8677042722702026], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004124.pt"}
{"idx": 4126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6yYd6Pq7Ic.mp4", "duration": 64.25, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the elf is getting his shoes shined.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the elf is getting his shoes shined.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.7198443412780762, -0.5953307151794434], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7198443412780762, -0.5953307151794434], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004126.pt"}
{"idx": 4128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6yYd6Pq7Ic.mp4", "duration": 64.25, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the elf congratulates the coffee shop staff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the elf congratulates the coffee shop staff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 36.0], "base_se_norm": [-0.3151751160621643, 0.12062251567840576], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.3151751160621643, -0.1595330834388733], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004128.pt"}
{"idx": 4130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r7MWpMelEyg.mp4", "duration": 114.58113333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the girl puts the transparent rattle into the metal bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the girl puts the transparent rattle into the metal bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [-0.24944013357162476, -0.16216570138931274], "gt_se_raw": [31.0, 42.0], "gt_se_norm": [-0.458898663520813, -0.26689499616622925], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004130.pt"}
{"idx": 4132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r7MWpMelEyg.mp4", "duration": 114.58113333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the girl is playing with the stacking ring toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the girl is playing with the stacking ring toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 87.0], "base_se_norm": [0.3091161251068115, 0.518574595451355], "gt_se_raw": [70.0, 81.0], "gt_se_norm": [0.2218416929244995, 0.4138453006744385], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004132.pt"}
{"idx": 4134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/egkrxkiUnoo.mp4", "duration": 150.01653333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the bus drive past the man on the bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bus drive past the man on the bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, -0.853349506855011], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, -0.8666813373565674], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004134.pt"}
{"idx": 4136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/egkrxkiUnoo.mp4", "duration": 150.01653333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man says the line, 'Life was like a box of chocolates'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man says the line, 'Life was like a box of chocolates'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 96.0], "base_se_norm": [0.22653138637542725, 0.2798588275909424], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.41339802742004395, -0.3734024167060852], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004136.pt"}
{"idx": 4138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/egkrxkiUnoo.mp4", "duration": 150.01653333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the boy walk in his leg braces for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy walk in his leg braces for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 137.0], "base_se_norm": [0.6398191452026367, 0.8264652490615845], "gt_se_raw": [123.0, 127.0], "gt_se_norm": [0.6398191452026367, 0.6931465864181519], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004138.pt"}
{"idx": 4140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qi2m4V21bw4.mp4", "duration": 179.0, "prompt_text": "Please find the visual event described by the sentence 'At what time does Mr. Bean cook pasta in the bathtub?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does Mr. Bean cook pasta in the bathtub?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 72.0], "base_se_norm": [-0.4189944267272949, -0.1955307126045227], "gt_se_raw": [50.0, 72.0], "gt_se_norm": [-0.4413408041000366, -0.1955307126045227], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004140.pt"}
{"idx": 4142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qi2m4V21bw4.mp4", "duration": 179.0, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where Mr. Bean is brushing his teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where Mr. Bean is brushing his teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 103.0], "base_se_norm": [0.061452507972717285, 0.15083801746368408], "gt_se_raw": [95.0, 104.0], "gt_se_norm": [0.061452507972717285, 0.16201114654541016], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004142.pt"}
{"idx": 4144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LKFuXETZUsI.mp4", "duration": 154.40425, "prompt_text": "Please find the visual event described by the sentence 'When can we see Maui shapeshift into a bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Maui shapeshift into a bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 50.0], "base_se_norm": [-0.4171144962310791, -0.3523494601249695], "gt_se_raw": [46.0, 47.0], "gt_se_norm": [-0.4041614532470703, -0.3912084698677063], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004144.pt"}
{"idx": 4146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LKFuXETZUsI.mp4", "duration": 154.40425, "prompt_text": "Please find the visual event described by the sentence 'At what time does the ocean splash Maui?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the ocean splash Maui?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 72.0], "base_se_norm": [-0.09328919649124146, -0.06738317012786865], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [-0.09328919649124146, -0.06738317012786865], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004146.pt"}
{"idx": 4148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LKFuXETZUsI.mp4", "duration": 154.40425, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a girl is giving water to a pig.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a girl is giving water to a pig.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.9481879472732544, -0.9222819209098816], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.9481879472732544, -0.9352349638938904], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004148.pt"}
{"idx": 4150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PjCZVeR0do8.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the narrator says that disabled people are employed at almost half the rate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the narrator says that disabled people are employed at almost half the rate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9031662940979004, -0.6126651763916016], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.1284967064857483, 0.355671763420105], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004150.pt"}
{"idx": 4152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WNHJnUIGQ_U.mp4", "duration": 95.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman take a green toothbrush out of the red container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman take a green toothbrush out of the red container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.3715083599090576, -0.28770947456359863], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [0.2779330015182495, 0.34078216552734375], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004152.pt"}
{"idx": 4154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WNHJnUIGQ_U.mp4", "duration": 95.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the girl's hand to hold the brown toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the girl's hand to hold the brown toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.6229050159454346, -0.5181564092636108], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [0.5293296575546265, 0.6340781450271606], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004154.pt"}
{"idx": 4156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_lmKuKrsKRA.mp4", "duration": 82.457375, "prompt_text": "Please find the visual event described by the sentence 'Show me when the pancakes and syrup materialize on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the pancakes and syrup materialize on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9272351264953613, -0.8059603571891785], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.9272351264953613, -0.9029802083969116], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004156.pt"}
{"idx": 4158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_lmKuKrsKRA.mp4", "duration": 82.457375, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman attempt to force-feed the man with a spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman attempt to force-feed the man with a spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.1510765552520752, -0.07831168174743652], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.1510765552520752, -0.07831168174743652], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004158.pt"}
{"idx": 4160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_lmKuKrsKRA.mp4", "duration": 82.457375, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man is eating food directly from a can with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man is eating food directly from a can with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 59.0], "base_se_norm": [0.11572802066802979, 0.4310424327850342], "gt_se_raw": [47.0, 51.0], "gt_se_norm": [0.1399829387664795, 0.23700273036956787], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004160.pt"}
{"idx": 4162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ppyYdn2nPoU.mp4", "duration": 124.72472472472472, "prompt_text": "Please find the visual event described by the sentence 'When are the students engaging in decorating activities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the students engaging in decorating activities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 39.0], "base_se_norm": [-0.6953290700912476, -0.37462276220321655], "gt_se_raw": [19.0, 39.0], "gt_se_norm": [-0.6953290700912476, -0.37462276220321655], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004162.pt"}
{"idx": 4164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ppyYdn2nPoU.mp4", "duration": 124.72472472472472, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where students are showing off their decorated canes in a fashion show.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where students are showing off their decorated canes in a fashion show.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 88.0], "base_se_norm": [0.13850724697113037, 0.41110754013061523], "gt_se_raw": [71.0, 88.0], "gt_se_norm": [0.13850724697113037, 0.41110754013061523], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004164.pt"}
{"idx": 4166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/16K6BDTn9iI.mp4", "duration": 28.72, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman with long red hair is standing in a snowy courtyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman with long red hair is standing in a snowy courtyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.44289690256118774, -0.30362117290496826], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [-0.44289690256118774, -0.3732590675354004], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004166.pt"}
{"idx": 4168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/16K6BDTn9iI.mp4", "duration": 28.72, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a blonde woman is sitting next to a large black dragon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a blonde woman is sitting next to a large black dragon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [0.04456830024719238, 0.18384408950805664], "gt_se_raw": [16.0, 17.0], "gt_se_norm": [0.11420619487762451, 0.18384408950805664], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004168.pt"}
{"idx": 4170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/16K6BDTn9iI.mp4", "duration": 28.72, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where giants are riding mammoths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where giants are riding mammoths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [0.4623955488204956, 0.6016713380813599], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [0.5320334434509277, 0.6016713380813599], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004170.pt"}
{"idx": 4172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XbKhZYOvx8.mp4", "duration": 83.08, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a group of people are protesting in front of a brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a group of people are protesting in front of a brick building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [-0.4222436547279358, -0.3740972876548767], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.4222436547279358, -0.3740972876548767], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004172.pt"}
{"idx": 4174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XbKhZYOvx8.mp4", "duration": 83.08, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man hold up a sign that says 'I AM NOT YOUR INSPIRATION PORN'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man hold up a sign that says 'I AM NOT YOUR INSPIRATION PORN'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.42031776905059814, 0.5166105031967163], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.42031776905059814, 0.492537260055542], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004174.pt"}
{"idx": 4176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xqZQ9KM_LjY.mp4", "duration": 198.7986, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the white shirt place the pan on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the white shirt place the pan on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.9195165634155273, -0.8893352150917053], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.9195165634155273, -0.8993956446647644], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004176.pt"}
{"idx": 4178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xqZQ9KM_LjY.mp4", "duration": 198.7986, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man in the white shirt presents the Super Nintendo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man in the white shirt presents the Super Nintendo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 105.0], "base_se_norm": [0.02616417407989502, 0.056345462799072266], "gt_se_raw": [103.0, 105.0], "gt_se_norm": [0.0362246036529541, 0.056345462799072266], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004178.pt"}
{"idx": 4180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xqZQ9KM_LjY.mp4", "duration": 198.7986, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man in the pink shirt smashes an imaginary vase over the other man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man in the pink shirt smashes an imaginary vase over the other man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 154.0], "base_se_norm": [0.5191253423690796, 0.5493067502975464], "gt_se_raw": [150.0, 152.0], "gt_se_norm": [0.5090649127960205, 0.5291857719421387], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004180.pt"}
{"idx": 4182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jHe5vPlKgJA.mp4", "duration": 21.755066666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy with glasses turns his head to look at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy with glasses turns his head to look at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.5403369665145874, -0.26453912258148193], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.44840431213378906, -0.26453912258148193], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004182.pt"}
{"idx": 4184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jHe5vPlKgJA.mp4", "duration": 21.755066666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where one boy climbs onto the other's back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where one boy climbs onto the other's back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [0.3789891004562378, 0.8386522531509399], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.47092175483703613, 0.6547869443893433], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004184.pt"}
{"idx": 4186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2bpkd6hwH6U.mp4", "duration": 143.768625, "prompt_text": "Please find the visual event described by the sentence 'When are the woman's feet being chained?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman's feet being chained?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [-0.819153904914856, -0.79133141040802], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.819153904914856, -0.79133141040802], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004186.pt"}
{"idx": 4188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2bpkd6hwH6U.mp4", "duration": 143.768625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman struggle to unlock the chains on her feet underwater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman struggle to unlock the chains on her feet underwater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 47.0], "base_se_norm": [-0.5548403263092041, -0.34617167711257935], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.38790541887283325, -0.34617167711257935], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004188.pt"}
{"idx": 4190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2bpkd6hwH6U.mp4", "duration": 143.768625, "prompt_text": "Please find the visual event described by the sentence 'When does the water in the tank turn red?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the water in the tank turn red?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 98.0], "base_se_norm": [0.3076566457748413, 0.36330151557922363], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.3076566457748413, 0.34939026832580566], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004190.pt"}
{"idx": 4192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uidTVaoTjNY.mp4", "duration": 130.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the animation of various public transport methods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the animation of various public transport methods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 57.0], "base_se_norm": [-0.18690872192382812, -0.12554335594177246], "gt_se_raw": [53.0, 58.0], "gt_se_norm": [-0.18690872192382812, -0.11020201444625854], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004192.pt"}
{"idx": 4194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uidTVaoTjNY.mp4", "duration": 130.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a megaphone is used to represent awareness.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a megaphone is used to represent awareness.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 92.0], "base_se_norm": [0.30401432514190674, 0.4114036560058594], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.30401432514190674, 0.4114036560058594], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004194.pt"}
{"idx": 4196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mEBtnA4DIM4.mp4", "duration": 123.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the escalator suddenly stop for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the escalator suddenly stop for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [0.06594884395599365, 0.14670252799987793], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.8223418593406677, -0.7900403738021851], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004196.pt"}
{"idx": 4198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mEBtnA4DIM4.mp4", "duration": 123.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman shout for help?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman shout for help?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 48.0], "base_se_norm": [-0.3378196358680725, -0.22476446628570557], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.22476446628570557, -0.1924629807472229], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004198.pt"}
{"idx": 4200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mEBtnA4DIM4.mp4", "duration": 123.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person who came to help trips and falls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person who came to help trips and falls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 75.0], "base_se_norm": [0.16285324096679688, 0.21130549907684326], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.3728128671646118, 0.40511441230773926], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004200.pt"}
{"idx": 4202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V99PmrBPUqk.mp4", "duration": 161.28, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where there's a traffic jam on the highway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where there's a traffic jam on the highway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 21.0], "base_se_norm": [-0.838789701461792, -0.7395833730697632], "gt_se_raw": [13.0, 21.0], "gt_se_norm": [-0.838789701461792, -0.7395833730697632], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004202.pt"}
{"idx": 4204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V99PmrBPUqk.mp4", "duration": 161.28, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man in the wheelchair is getting into the concept car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man in the wheelchair is getting into the concept car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 104.0], "base_se_norm": [0.05406749248504639, 0.2896825075149536], "gt_se_raw": [91.0, 105.0], "gt_se_norm": [0.1284722089767456, 0.3020833730697632], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004204.pt"}
{"idx": 4206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsQMSvhMcMc.mp4", "duration": 124.39093333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the exterior of the museum building from a low angle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the exterior of the museum building from a low angle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 27.0], "base_se_norm": [-0.6623548269271851, -0.5658847093582153], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [-0.6623548269271851, -0.5658847093582153], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004206.pt"}
{"idx": 4208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsQMSvhMcMc.mp4", "duration": 124.39093333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man in a vest and glasses is talking in front of an exhibit wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man in a vest and glasses is talking in front of an exhibit wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 62.0], "base_se_norm": [-0.4211796522140503, -0.0031427741050720215], "gt_se_raw": [37.0, 52.0], "gt_se_norm": [-0.40510129928588867, -0.16392618417739868], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004208.pt"}
{"idx": 4210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsQMSvhMcMc.mp4", "duration": 124.39093333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range where the museum's hours and admission information is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range where the museum's hours and admission information is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 109.0], "base_se_norm": [0.6239125728607178, 0.7525393962860107], "gt_se_raw": [102.0, 109.0], "gt_se_norm": [0.6399909257888794, 0.7525393962860107], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004210.pt"}
{"idx": 4212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cda7mSowTEI.mp4", "duration": 75.50884217550885, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman demonstrate drawing techniques from an overhead perspective?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman demonstrate drawing techniques from an overhead perspective?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 30.0], "base_se_norm": [-0.5497215986251831, -0.2053910493850708], "gt_se_raw": [17.0, 31.0], "gt_se_norm": [-0.5497215986251831, -0.17890411615371704], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004212.pt"}
{"idx": 4214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cda7mSowTEI.mp4", "duration": 75.50884217550885, "prompt_text": "Please find the visual event described by the sentence 'Show the artwork featuring a cat named 'Luna'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the artwork featuring a cat named 'Luna'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [0.6421917676925659, 0.721652626991272], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [0.6421917676925659, 0.721652626991272], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004214.pt"}
{"idx": 4216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jhfLlamufKE.mp4", "duration": 56.72333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man have a smile on his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man have a smile on his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.5416347980499268, -0.43585824966430664], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.57689368724823, -0.4711170792579651], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004216.pt"}
{"idx": 4218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jhfLlamufKE.mp4", "duration": 56.72333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man looks down at his plate with a dejected expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man looks down at his plate with a dejected expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [0.12828350067138672, 0.23406004905700684], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [0.12828350067138672, 0.3045778274536133], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004218.pt"}
{"idx": 4220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jhfLlamufKE.mp4", "duration": 56.72333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the unopened can with 'FOOD' written on it in front of the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the unopened can with 'FOOD' written on it in front of the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 30.0], "base_se_norm": [-0.5063759684562683, 0.057765841484069824], "gt_se_raw": [0.0, 28.0], "gt_se_norm": [-1.0, -0.012751936912536621], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004220.pt"}
{"idx": 4222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g26mbST0YhU.mp4", "duration": 91.4914, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the boy in the red shirt lifts the front of the walking aid by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the boy in the red shirt lifts the front of the walking aid by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 72.0], "base_se_norm": [0.44275856018066406, 0.573918342590332], "gt_se_raw": [67.0, 70.0], "gt_se_norm": [0.4646185636520386, 0.5301984548568726], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004222.pt"}
{"idx": 4224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g26mbST0YhU.mp4", "duration": 91.4914, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the woman straightens the boy's hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the woman straightens the boy's hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.45350056886672974, -0.387920618057251], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [-0.4316405653953552, -0.4097806215286255], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004224.pt"}
{"idx": 4226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uNefa-pBVD0.mp4", "duration": 75.04163333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the artwork that looks like a balance scale with two golden necklaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the artwork that looks like a balance scale with two golden necklaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.5469181537628174, -0.4669625759124756], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.5469181537628174, -0.4669625759124756], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004226.pt"}
{"idx": 4228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uNefa-pBVD0.mp4", "duration": 75.04163333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is an empty basketball court with banners hanging from the balcony shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an empty basketball court with banners hanging from the balcony shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.06718450784683228, 0.14603054523468018], "gt_se_raw": [34.0, 42.0], "gt_se_norm": [-0.09383636713027954, 0.11937856674194336], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004228.pt"}
{"idx": 4230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uNefa-pBVD0.mp4", "duration": 75.04163333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the close-up shot of the words 'MY TEARS RUN LIKE ACID RAIN' carved into a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the close-up shot of the words 'MY TEARS RUN LIKE ACID RAIN' carved into a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [0.5191566944122314, 0.6524161100387573], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [0.5191566944122314, 0.6257641315460205], "start_token_span": [83, 87], "end_token_span": [89, 93], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004230.pt"}
{"idx": 4232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7cSIiKxEt4.mp4", "duration": 70.87078745412079, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a woman operating a radio broadcast console?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a woman operating a radio broadcast console?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 45.0], "base_se_norm": [-0.3791518211364746, 0.2699167728424072], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [-0.35093146562576294, -0.18160921335220337], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004232.pt"}
{"idx": 4234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7cSIiKxEt4.mp4", "duration": 70.87078745412079, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man with glasses and a red shirt is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man with glasses and a red shirt is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [0.4674593210220337, 0.5803408622741699], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [0.4674593210220337, 0.5803408622741699], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004234.pt"}
{"idx": 4236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zYVR5PDom8.mp4", "duration": 88.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the girl in the white top speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl in the white top speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 24.0], "base_se_norm": [-0.6381455063819885, -0.4572182297706604], "gt_se_raw": [16.0, 24.0], "gt_se_norm": [-0.6381455063819885, -0.4572182297706604], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004236.pt"}
{"idx": 4238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zYVR5PDom8.mp4", "duration": 88.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does one girl ask to borrow a purple crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one girl ask to borrow a purple crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.29890692234039307, -0.23105919361114502], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.29890692234039307, -0.23105919361114502], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004238.pt"}
{"idx": 4240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zYVR5PDom8.mp4", "duration": 88.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does a boy in a white polo shirt play with a paddle ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a boy in a white polo shirt play with a paddle ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 66.0], "base_se_norm": [0.3117225170135498, 0.49264979362487793], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [0.3117225170135498, 0.40218615531921387], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004240.pt"}
{"idx": 4242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ENaJcHwQEVs.mp4", "duration": 56.62323333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy rolls from his back to his side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy rolls from his back to his side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.7527516484260559, -0.4701821208000183], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [0.2009204626083374, 0.3068840503692627], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004242.pt"}
{"idx": 4244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ENaJcHwQEVs.mp4", "duration": 56.62323333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy lying on his back with his knees bent and feet touching the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy lying on his back with his knees bent and feet touching the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, -0.4701821208000183], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.7174304723739624], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004244.pt"}
{"idx": 4246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UGvlQOWyB2Y.mp4", "duration": 31.533333333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the mascot skate near the goal while waving the orange flag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the mascot skate near the goal while waving the orange flag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.04862576723098755, 0.3319239616394043], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.014799118041992188, 0.3319239616394043], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004246.pt"}
{"idx": 4248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UGvlQOWyB2Y.mp4", "duration": 31.533333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where individuals are skating on the ice holding blue flags.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where individuals are skating on the ice holding blue flags.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.5856237411499023, 0.7758985757827759], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.3023255467414856, -0.11205071210861206], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004248.pt"}
{"idx": 4250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDvBqq3v8bM.mp4", "duration": 360.89386666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where it shows a map of Africa with Cameroon highlighted.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where it shows a map of Africa with Cameroon highlighted.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [-0.7672445774078369, -0.7339937686920166], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.7617027759552002, -0.7339937686920166], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004250.pt"}
{"idx": 4252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDvBqq3v8bM.mp4", "duration": 360.89386666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman in a wheelchair going up a ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 333.0 - 338.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman in a wheelchair going up a ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 333.0 - 338.0 seconds.<|im_end|>\n", "base_se_raw": [333.0, 338.0], "base_se_norm": [0.8454179763793945, 0.8731268644332886], "gt_se_raw": [333.0, 338.0], "gt_se_norm": [0.8454179763793945, 0.8731268644332886], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004252.pt"}
{"idx": 4254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDvBqq3v8bM.mp4", "duration": 360.89386666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a man is using sign language in what appears to be a radio studio.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 309.0 - 317.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a man is using sign language in what appears to be a radio studio.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 309.0 - 317.0 seconds.<|im_end|>\n", "base_se_raw": [309.0, 317.0], "base_se_norm": [0.7124148607254028, 0.756749153137207], "gt_se_raw": [309.0, 317.0], "gt_se_norm": [0.7124148607254028, 0.756749153137207], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004254.pt"}
{"idx": 4256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7VS1wPeWqAA.mp4", "duration": 83.5835, "prompt_text": "Please find the visual event described by the sentence 'Show the interaction between the man and the two women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the interaction between the man and the two women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 48.0], "base_se_norm": [-0.8564310073852539, 0.14855194091796875], "gt_se_raw": [7.0, 48.0], "gt_se_norm": [-0.8325028419494629, 0.14855194091796875], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004256.pt"}
{"idx": 4258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7VS1wPeWqAA.mp4", "duration": 83.5835, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman in the pink sweater leaves the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman in the pink sweater leaves the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [0.004982948303222656, 0.14855194091796875], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [0.0767674446105957, 0.14855194091796875], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004258.pt"}
{"idx": 4260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM7Mh9T_XE.mp4", "duration": 142.35053333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the woman applying lipstick in front of the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman applying lipstick in front of the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.9157010316848755], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.943800687789917], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004260.pt"}
{"idx": 4262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM7Mh9T_XE.mp4", "duration": 142.35053333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the man is drinking orange juice from the carton.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the man is drinking orange juice from the carton.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 33.0], "base_se_norm": [-0.5644554495811462, -0.5363558530807495], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.578505277633667, -0.5504056215286255], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004262.pt"}
{"idx": 4264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM7Mh9T_XE.mp4", "duration": 142.35053333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when all three people are sitting at the dining table and talking on their phones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when all three people are sitting at the dining table and talking on their phones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 130.0], "base_se_norm": [0.756227970123291, 0.82647705078125], "gt_se_raw": [124.0, 128.0], "gt_se_norm": [0.742178201675415, 0.7983773946762085], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004264.pt"}
{"idx": 4266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cHUNbTfzOr4.mp4", "duration": 72.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman talking about the successes of the program?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman talking about the successes of the program?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 52.0], "base_se_norm": [-0.4480220675468445, 0.4351426362991333], "gt_se_raw": [27.0, 52.0], "gt_se_norm": [-0.2548298239707947, 0.4351426362991333], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004266.pt"}
{"idx": 4268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cHUNbTfzOr4.mp4", "duration": 72.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame in which the woman asks viewers to call if they have any questions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame in which the woman asks viewers to call if they have any questions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 63.0], "base_se_norm": [0.4351426362991333, 0.7387304306030273], "gt_se_raw": [53.0, 61.0], "gt_se_norm": [0.462741494178772, 0.6835325956344604], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004268.pt"}
{"idx": 4270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kgDjwR0xoQg.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman point to the limbic system on the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman point to the limbic system on the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [-0.7944129109382629, -0.6573548316955566], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7944129109382629, -0.7030408382415771], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004270.pt"}
{"idx": 4272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kgDjwR0xoQg.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention the example of 'grandma's kitchen'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention the example of 'grandma's kitchen'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.3147096633911133, -0.2233375906944275], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.3147096633911133, -0.2690236568450928], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004272.pt"}
{"idx": 4274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kgDjwR0xoQg.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is talking about the smell of disinfectants in a hospital.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is talking about the smell of disinfectants in a hospital.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 77.0], "base_se_norm": [0.3934236764907837, 0.7589118480682373], "gt_se_raw": [65.0, 69.0], "gt_se_norm": [0.4847956895828247, 0.5761678218841553], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004274.pt"}
{"idx": 4276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wC1ZQNR_1N4.mp4", "duration": 192.5, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man gives a tip to the employee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man gives a tip to the employee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 27.0], "base_se_norm": [-0.8129870295524597, -0.7194805145263672], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.8129870295524597, -0.7714285850524902], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004276.pt"}
{"idx": 4278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wC1ZQNR_1N4.mp4", "duration": 192.5, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man starts drinking the slushie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man starts drinking the slushie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 136.0], "base_se_norm": [0.29870128631591797, 0.41298699378967285], "gt_se_raw": [125.0, 130.0], "gt_se_norm": [0.29870128631591797, 0.350649356842041], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004278.pt"}
{"idx": 4280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OxrBik16Hzg.mp4", "duration": 95.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a man pouring liquid over another man who is sitting on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man pouring liquid over another man who is sitting on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.811188817024231, -0.7062937021255493], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.8321678638458252, -0.7272727489471436], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004280.pt"}
{"idx": 4282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OxrBik16Hzg.mp4", "duration": 95.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 71.0], "base_se_norm": [-0.1818181872367859, 0.4895104169845581], "gt_se_raw": [40.0, 45.0], "gt_se_norm": [-0.16083920001983643, -0.055944085121154785], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004282.pt"}
{"idx": 4284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OxrBik16Hzg.mp4", "duration": 95.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the text 'WHILE WANDERING, WE LEAD A LIFE FULL OF PAIN AND SUFFERING...'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the text 'WHILE WANDERING, WE LEAD A LIFE FULL OF PAIN AND SUFFERING...'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 84.0], "base_se_norm": [0.6783215999603271, 0.7622376680374146], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.6993006467819214, 0.7622376680374146], "start_token_span": [82, 86], "end_token_span": [88, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004284.pt"}
{"idx": 4286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5VmfgTfuYGE.mp4", "duration": 48.757041666666666, "prompt_text": "Please find the visual event described by the sentence 'When is a woman pointing at a whiteboard and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman pointing at a whiteboard and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 22.0], "base_se_norm": [-0.1796056628227234, -0.09756624698638916], "gt_se_raw": [21.0, 22.0], "gt_se_norm": [-0.13858598470687866, -0.09756624698638916], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004286.pt"}
{"idx": 4288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5VmfgTfuYGE.mp4", "duration": 48.757041666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman with curly hair speak into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman with curly hair speak into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 42.0], "base_se_norm": [0.6407886743545532, 0.7228280305862427], "gt_se_raw": [41.0, 42.0], "gt_se_norm": [0.681808352470398, 0.7228280305862427], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004288.pt"}
{"idx": 4290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zCdVxICvEGg.mp4", "duration": 118.8187, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a black top doing sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a black top doing sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.8148439526557922, -0.7643468379974365], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.8148439526557922, -0.7643468379974365], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004290.pt"}
{"idx": 4292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zCdVxICvEGg.mp4", "duration": 118.8187, "prompt_text": "Please find the visual event described by the sentence 'At what point do the audience members clap for the performer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the audience members clap for the performer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.09105217456817627, -0.04055505990982056], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.09105217456817627, -0.04055505990982056], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004292.pt"}
{"idx": 4294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zCdVxICvEGg.mp4", "duration": 118.8187, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman in an orange top is teaching a group of people in black shirts to raise their fists.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman in an orange top is teaching a group of people in black shirts to raise their fists.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 116.0], "base_se_norm": [0.8347280025482178, 0.9525545835494995], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [0.8515603542327881, 0.8852250576019287], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004294.pt"}
{"idx": 4296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V3wLiAmIrGk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When are two men inspecting the tires of a large white trailer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men inspecting the tires of a large white trailer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.5414421558380127, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004296.pt"}
{"idx": 4298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V3wLiAmIrGk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a plaid shirt shown sitting inside the truck's cab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a plaid shirt shown sitting inside the truck's cab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004298.pt"}
{"idx": 4300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cZdO2e8K29o.mp4", "duration": 190.2901, "prompt_text": "Please find the visual event described by the sentence 'When are the five finished butterflies of different colors displayed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the five finished butterflies of different colors displayed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9684692025184631], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9579589366912842], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004300.pt"}
{"idx": 4302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cZdO2e8K29o.mp4", "duration": 190.2901, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the paper is collapsed from a square into a triangular base.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the paper is collapsed from a square into a triangular base.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 85.0], "base_se_norm": [-0.3168325424194336, -0.10662722587585449], "gt_se_raw": [66.0, 76.0], "gt_se_norm": [-0.30632227659225464, -0.2012196183204651], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004302.pt"}
{"idx": 4304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cZdO2e8K29o.mp4", "duration": 190.2901, "prompt_text": "Please find the visual event described by the sentence 'When is the model folded in half along its central axis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the model folded in half along its central axis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 136.0], "base_se_norm": [0.2822526693344116, 0.42939651012420654], "gt_se_raw": [138.0, 146.0], "gt_se_norm": [0.45041704177856445, 0.5344991683959961], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004304.pt"}
{"idx": 4306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F0M7n8DGjBk.mp4", "duration": 180.26341666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man dressed in black is walking in the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man dressed in black is walking in the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 108.0], "base_se_norm": [0.03182339668273926, 0.19824647903442383], "gt_se_raw": [94.0, 103.0], "gt_se_norm": [0.04291820526123047, 0.14277207851409912], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004306.pt"}
{"idx": 4308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F0M7n8DGjBk.mp4", "duration": 180.26341666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man and a woman are walking together holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man and a woman are walking together holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 149.0], "base_se_norm": [0.6309466361999512, 0.6531363725662231], "gt_se_raw": [147.0, 149.0], "gt_se_norm": [0.6309466361999512, 0.6531363725662231], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004308.pt"}
{"idx": 4310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YPFGT4ecnIU.mp4", "duration": 40.6, "prompt_text": "Please find the visual event described by the sentence 'When does the girl run from the kitchen island towards the living room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl run from the kitchen island towards the living room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.6551724076271057], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9014778137207031], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004310.pt"}
{"idx": 4312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YPFGT4ecnIU.mp4", "duration": 40.6, "prompt_text": "Please find the visual event described by the sentence 'At what point is there a close-up of a woman's face wearing glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there a close-up of a woman's face wearing glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.1330050230026245, 0.2807881832122803], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.1330050230026245, 0.23152709007263184], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004312.pt"}
{"idx": 4314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YPFGT4ecnIU.mp4", "duration": 40.6, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a voice says 'Stop running'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a voice says 'Stop running'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [0.3300492763519287, 0.477832555770874], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.8226602077484131, 0.92118239402771], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004314.pt"}
{"idx": 4316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w28U-Js6g1Q.mp4", "duration": 203.08, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman tells the man she is pregnant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman tells the man she is pregnant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 63.0], "base_se_norm": [-0.4583415389060974, -0.3795548677444458], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [-0.44849318265914917, -0.40909987688064575], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004316.pt"}
{"idx": 4318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w28U-Js6g1Q.mp4", "duration": 203.08, "prompt_text": "Please find the visual event described by the sentence 'Show the woman tearing up the sheet music and throwing it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 152.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman tearing up the sheet music and throwing it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 152.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 152.0], "base_se_norm": [0.4181603193283081, 0.4969470500946045], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [0.3393735885620117, 0.3590703010559082], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004318.pt"}
{"idx": 4320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K23OHlnu0e0.mp4", "duration": 65.99926666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman holding a yellow brain-shaped object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman holding a yellow brain-shaped object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 65.0], "base_se_norm": [-1.0, 0.9697188138961792], "gt_se_raw": [0.0, 65.0], "gt_se_norm": [-1.0, 0.9697188138961792], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004320.pt"}
{"idx": 4322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K23OHlnu0e0.mp4", "duration": 65.99926666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the speaker explains what the 'thinking brain' is responsible for.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the speaker explains what the 'thinking brain' is responsible for.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 48.0], "base_se_norm": [-0.09089899063110352, 0.45456159114837646], "gt_se_raw": [26.0, 34.0], "gt_se_norm": [-0.21211248636245728, 0.03031444549560547], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004322.pt"}
{"idx": 4324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K23OHlnu0e0.mp4", "duration": 65.99926666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker begin to list the different senses like vision and hearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker begin to list the different senses like vision and hearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 20.0], "base_se_norm": [-0.696966290473938, -0.39393264055252075], "gt_se_raw": [46.0, 50.0], "gt_se_norm": [0.393954873085022, 0.515168309211731], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004324.pt"}
{"idx": 4326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0GtT-vN33I.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are two men working together on the wheel assembly of a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two men working together on the wheel assembly of a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004326.pt"}
{"idx": 4328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0GtT-vN33I.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a young man in a plaid jacket is using a power tool on a car's frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a young man in a plaid jacket is using a power tool on a car's frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004328.pt"}
{"idx": 4330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzzjgBAaWZw.mp4", "duration": 73.2732, "prompt_text": "Please find the visual event described by the sentence 'When is the cat seen walking slowly from the hallway into the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cat seen walking slowly from the hallway into the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 15.0], "base_se_norm": [-0.9727048873901367, -0.5905733704566956], "gt_se_raw": [1.0, 10.0], "gt_se_norm": [-0.9727048873901367, -0.7270488739013672], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004330.pt"}
{"idx": 4332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzzjgBAaWZw.mp4", "duration": 73.2732, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cat gets very close, and its eyes are wide open before attacking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cat gets very close, and its eyes are wide open before attacking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 68.0], "base_se_norm": [0.665001630783081, 0.856067419052124], "gt_se_raw": [61.0, 67.0], "gt_se_norm": [0.665001630783081, 0.8287723064422607], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004332.pt"}
{"idx": 4334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzzjgBAaWZw.mp4", "duration": 73.2732, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the cat is seen running away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the cat is seen running away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [0.856067419052124, 0.9379527568817139], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [0.8833625316619873, 0.9379527568817139], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004334.pt"}
{"idx": 4336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahSKGi_21xg.mp4", "duration": 163.56356356356358, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman with short blonde hair and a black top is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman with short blonde hair and a black top is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 93.0], "base_se_norm": [0.07603418827056885, 0.13717257976531982], "gt_se_raw": [21.0, 31.0], "gt_se_norm": [-0.7432191371917725, -0.6209424734115601], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004336.pt"}
{"idx": 4338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahSKGi_21xg.mp4", "duration": 163.56356356356358, "prompt_text": "Please find the visual event described by the sentence 'Find the segment featuring the man with long dark hair and a white hoodie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment featuring the man with long dark hair and a white hoodie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 68.0], "base_se_norm": [-0.2541126012802124, -0.1685190200805664], "gt_se_raw": [62.0, 68.0], "gt_se_norm": [-0.24188494682312012, -0.1685190200805664], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004338.pt"}
{"idx": 4340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZFpnzOGnZBM.mp4", "duration": 128.39493333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a shot of the beach and ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a shot of the beach and ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.8909614086151123, -0.844230592250824], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8909614086151123, -0.844230592250824], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004340.pt"}
{"idx": 4342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZFpnzOGnZBM.mp4", "duration": 128.39493333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip of a person using a wheelchair in the snow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip of a person using a wheelchair in the snow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [-0.33019161224365234, -0.2523069381713867], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.33019161224365234, -0.2523069381713867], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004342.pt"}
{"idx": 4344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZFpnzOGnZBM.mp4", "duration": 128.39493333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene that displays a collection of eight photos showcasing people using the wheelchair in various environments.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene that displays a collection of eight photos showcasing people using the wheelchair in various environments.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [0.05923187732696533, 0.13711655139923096], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [0.05923187732696533, 0.13711655139923096], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004344.pt"}
{"idx": 4346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3aZadizk_Io.mp4", "duration": 162.12863333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the lion look at his own reflection in the pool of water, before it is disturbed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the lion look at his own reflection in the pool of water, before it is disturbed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 73.0], "base_se_norm": [-0.1981675624847412, -0.09948045015335083], "gt_se_raw": [70.0, 79.0], "gt_se_norm": [-0.136488139629364, -0.02546513080596924], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004346.pt"}
{"idx": 4348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3aZadizk_Io.mp4", "duration": 162.12863333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when the monkey puts his finger to his lips to signal for quiet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when the monkey puts his finger to his lips to signal for quiet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.5312364101409912, -0.4818928837776184], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.5065646171569824, -0.4818928837776184], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004348.pt"}
{"idx": 4350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oyv6hjQr8MA.mp4", "duration": 124.75796666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the cost breakdown of the Standard Meal Plan shown with a pie chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cost breakdown of the Standard Meal Plan shown with a pie chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 29.0], "base_se_norm": [-0.647317111492157, -0.5350998640060425], "gt_se_raw": [22.0, 29.0], "gt_se_norm": [-0.647317111492157, -0.5350998640060425], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004350.pt"}
{"idx": 4352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oyv6hjQr8MA.mp4", "duration": 124.75796666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where all three meal plans are compared in a table format.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where all three meal plans are compared in a table format.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 87.0], "base_se_norm": [-0.07019966840744019, 0.3947005271911621], "gt_se_raw": [58.0, 87.0], "gt_se_norm": [-0.07019966840744019, 0.3947005271911621], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004352.pt"}
{"idx": 4354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oyv6hjQr8MA.mp4", "duration": 124.75796666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that mentions the deadlines for changing meal plans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that mentions the deadlines for changing meal plans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 116.0], "base_se_norm": [0.7794455289840698, 0.8596006631851196], "gt_se_raw": [111.0, 116.0], "gt_se_norm": [0.7794455289840698, 0.8596006631851196], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004354.pt"}
{"idx": 4356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sBdqOWSZ56w.mp4", "duration": 49.3, "prompt_text": "Please find the visual event described by the sentence 'At what point are the hands shown holding the yarn in the initial 'slingshot' position before the needle is used?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the hands shown holding the yarn in the initial 'slingshot' position before the needle is used?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.4320486783981323, -0.1480730175971985], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.3914807438850403, -0.2697768807411194], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004356.pt"}
{"idx": 4358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sBdqOWSZ56w.mp4", "duration": 49.3, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the completed row of cast-on stitches is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the completed row of cast-on stitches is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [0.501014232635498, 0.6632859706878662], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [0.501014232635498, 0.6632859706878662], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004358.pt"}
{"idx": 4360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ineZXLbL7s8.mp4", "duration": 100.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the part about using a burning newspaper for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part about using a burning newspaper for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.19999998807907104, 0.039999961853027344], "gt_se_raw": [41.0, 52.0], "gt_se_norm": [-0.18000000715255737, 0.039999961853027344], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004360.pt"}
{"idx": 4362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ineZXLbL7s8.mp4", "duration": 100.0, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where someone is creating a long exposure shot of a tram.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where someone is creating a long exposure shot of a tram.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.4599999785423279, -0.36000001430511475], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.4599999785423279, -0.36000001430511475], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004362.pt"}
{"idx": 4364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ineZXLbL7s8.mp4", "duration": 100.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a camera is made to look like it's levitating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a camera is made to look like it's levitating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 71.0], "base_se_norm": [0.24000000953674316, 0.41999995708465576], "gt_se_raw": [62.0, 71.0], "gt_se_norm": [0.24000000953674316, 0.41999995708465576], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004364.pt"}
{"idx": 4366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zMqQNCmFQE.mp4", "duration": 100.16683350016683, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a pile of white plates are smashed on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a pile of white plates are smashed on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 45.0], "base_se_norm": [-0.3011658787727356, -0.10149896144866943], "gt_se_raw": [36.0, 45.0], "gt_se_norm": [-0.28119921684265137, -0.10149896144866943], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004366.pt"}
{"idx": 4368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zMqQNCmFQE.mp4", "duration": 100.16683350016683, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the can, filled with eggs, explodes in a burst of fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the can, filled with eggs, explodes in a burst of fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [0.4775350093841553, 0.5574017763137817], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [0.5174683332443237, 0.5574017763137817], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004368.pt"}
{"idx": 4370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CIJjyEK9jB8.mp4", "duration": 189.95643333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman using an electronic braille device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman using an electronic braille device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7367817163467407, -0.7051955461502075], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.7262530326843262, -0.7051955461502075], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004370.pt"}
{"idx": 4372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CIJjyEK9jB8.mp4", "duration": 189.95643333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman with purple hair is giving a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman with purple hair is giving a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 146.0], "base_se_norm": [0.2950338125228882, 0.5371946096420288], "gt_se_raw": [123.0, 128.0], "gt_se_norm": [0.2950338125228882, 0.34767746925354004], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004372.pt"}
{"idx": 4374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CIJjyEK9jB8.mp4", "duration": 189.95643333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman read a Braille book with her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman read a Braille book with her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.4314485788345337, -0.3366900086402893], "gt_se_raw": [48.0, 63.0], "gt_se_norm": [-0.4946209788322449, -0.3366900086402893], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004374.pt"}
{"idx": 4376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1a7wwE952_E.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a black shirt looking at a diagnostic computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a black shirt looking at a diagnostic computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.015223503112792969, 0.3786870241165161], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004376.pt"}
{"idx": 4378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1a7wwE952_E.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a blue car elevated on a lift?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a blue car elevated on a lift?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47478586435317993, -0.08087527751922607], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004378.pt"}
{"idx": 4380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-7-quPaRol8.mp4", "duration": 187.12026666666668, "prompt_text": "Please find the visual event described by the sentence 'When is Dr. Arielle Silverman introduced as the first person being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Dr. Arielle Silverman introduced as the first person being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 25.0], "base_se_norm": [-0.7862337231636047, -0.7327921390533447], "gt_se_raw": [20.0, 26.0], "gt_se_norm": [-0.7862337231636047, -0.7221038341522217], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004380.pt"}
{"idx": 4382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-7-quPaRol8.mp4", "duration": 187.12026666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention a quote from Elissa Nadworny/NPR?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention a quote from Elissa Nadworny/NPR?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 74.0], "base_se_norm": [-0.25181812047958374, -0.20906484127044678], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [-0.2625064253807068, -0.20906484127044678], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004382.pt"}
{"idx": 4384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-7-quPaRol8.mp4", "duration": 187.12026666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where an image of many chairs appears with the text 'Not like this'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where an image of many chairs appears with the text 'Not like this'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 128.0], "base_se_norm": [0.33603906631469727, 0.36810410022735596], "gt_se_raw": [126.0, 129.0], "gt_se_norm": [0.3467273712158203, 0.378792405128479], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004384.pt"}
{"idx": 4386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aKIXMD3NQIA.mp4", "duration": 182.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where two women are communicating with each other using sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where two women are communicating with each other using sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 68.0], "base_se_norm": [-0.4642857313156128, -0.2565597891807556], "gt_se_raw": [50.0, 68.0], "gt_se_norm": [-0.45335274934768677, -0.2565597891807556], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004386.pt"}
{"idx": 4388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aKIXMD3NQIA.mp4", "duration": 182.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is using a brush on the girl's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is using a brush on the girl's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 122.0], "base_se_norm": [-0.23469388484954834, 0.33381927013397217], "gt_se_raw": [134.0, 139.0], "gt_se_norm": [0.46501457691192627, 0.5196793079376221], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004388.pt"}
{"idx": 4390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nwB2P3g1upM.mp4", "duration": 484.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of calculating 69 minus 56.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of calculating 69 minus 56.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 163.0], "base_se_norm": [-0.44283944368362427, -0.3272802233695984], "gt_se_raw": [52.0, 85.0], "gt_se_norm": [-0.7853900194168091, -0.6491951942443848], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004390.pt"}
{"idx": 4392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nwB2P3g1upM.mp4", "duration": 484.6, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the number 32 is set on the abacus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the number 32 is set on the abacus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 142.0], "base_se_norm": [-0.44283944368362427, -0.4139496684074402], "gt_se_raw": [94.0, 98.0], "gt_se_norm": [-0.6120511889457703, -0.5955427289009094], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004392.pt"}
{"idx": 4394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nwB2P3g1upM.mp4", "duration": 484.6, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for subtracting 13 from 64?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for subtracting 13 from 64?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 64.0], "base_se_norm": [-0.9463474750518799, -0.7358646392822266], "gt_se_raw": [141.0, 147.0], "gt_se_norm": [-0.418076753616333, -0.39331406354904175], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004394.pt"}
{"idx": 4396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oPwrodxghrw.mp4", "duration": 8.875533333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the brown suit speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the brown suit speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, 0.802708625793457], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.7746614217758179, -0.5493228435516357], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004396.pt"}
{"idx": 4398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oPwrodxghrw.mp4", "duration": 8.875533333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a 'FIRE HOSE' sign is visible in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a 'FIRE HOSE' sign is visible in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, 0.802708625793457], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.802708625793457], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004398.pt"}
{"idx": 4400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ycvwnsZ3ZCo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man in a lab coat is carefully pouring a white powder onto a digital scale.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man in a lab coat is carefully pouring a white powder onto a digital scale.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.606520414352417], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004400.pt"}
{"idx": 4402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ycvwnsZ3ZCo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a scientist is observing a sample through a microscope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a scientist is observing a sample through a microscope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004402.pt"}
{"idx": 4404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ycvwnsZ3ZCo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman in a lab coat smiles at the camera with the text 'YOU'VE GOT THIS!' on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman in a lab coat smiles at the camera with the text 'YOU'VE GOT THIS!' on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.508338451385498, 0.8362381458282471], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004404.pt"}
{"idx": 4406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rj07zM_GLeY.mp4", "duration": 51.38466666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy first put the wipe on his own face and laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy first put the wipe on his own face and laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [0.2844299077987671, 0.4790405035018921], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [0.2844299077987671, 0.4011962413787842], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004406.pt"}
{"idx": 4408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rj07zM_GLeY.mp4", "duration": 51.38466666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy takes the wipe again and holds it to his face, prompting the woman to laugh.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy takes the wipe again and holds it to his face, prompting the woman to laugh.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [0.2844299077987671, 0.5568846464157104], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [0.8682615756988525, 0.9850279092788696], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004408.pt"}
{"idx": 4410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gPdgsPTNhac.mp4", "duration": 141.57476666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a rocket launch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a rocket launch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 13.0], "base_se_norm": [-0.915239155292511, -0.8163514733314514], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.915239155292511, -0.8304782509803772], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004410.pt"}
{"idx": 4412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gPdgsPTNhac.mp4", "duration": 141.57476666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see Saturn's rings up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see Saturn's rings up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 33.0], "base_se_norm": [-0.6892101764678955, -0.5338152647018433], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.32191306352615356, -0.27953261137008667], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004412.pt"}
{"idx": 4414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gPdgsPTNhac.mp4", "duration": 141.57476666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the spacecraft plunging into the planet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the spacecraft plunging into the planet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 111.0], "base_se_norm": [0.44093477725982666, 0.5680760145187378], "gt_se_raw": [99.0, 105.0], "gt_se_norm": [0.39855432510375977, 0.483315110206604], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004414.pt"}
{"idx": 4416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txy3Joh7DZ4.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the instructor is guiding the boy's hand to draw on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the instructor is guiding the boy's hand to draw on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 34.0], "base_se_norm": [-0.5412474870681763, -0.1790744662284851], "gt_se_raw": [20.0, 34.0], "gt_se_norm": [-0.5171025991439819, -0.1790744662284851], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004416.pt"}
{"idx": 4418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txy3Joh7DZ4.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy points to the drawing by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy points to the drawing by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [0.18309855461120605, 0.35211265087127686], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [0.18309855461120605, 0.23138833045959473], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004418.pt"}
{"idx": 4420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G64wuf-rHoo.mp4", "duration": 46.046, "prompt_text": "Please find the visual event described by the sentence 'When are the people taking a group photo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the people taking a group photo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 37.0], "base_se_norm": [0.5202188491821289, 0.607088565826416], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.5202188491821289, 0.607088565826416], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004420.pt"}
{"idx": 4422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G64wuf-rHoo.mp4", "duration": 46.046, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a sign for the 'CSU Future Teacher Expo 6th Floor'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a sign for the 'CSU Future Teacher Expo 6th Floor'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [0.607088565826416, 0.7373930215835571], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.607088565826416, 0.6939581632614136], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004422.pt"}
{"idx": 4424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G64wuf-rHoo.mp4", "duration": 46.046, "prompt_text": "Please find the visual event described by the sentence 'When can we see three women sitting on a panel in front of a green wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three women sitting on a panel in front of a green wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.8696955442428589, -0.7393910884857178], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8696955442428589, -0.7828258872032166], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004424.pt"}
{"idx": 4426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3jb5aI4ejo.mp4", "duration": 298.3814111111111, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of her washing her face with a facial cloth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of her washing her face with a facial cloth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 82.0], "base_se_norm": [-0.6112358570098877, -0.45036792755126953], "gt_se_raw": [49.0, 83.0], "gt_se_norm": [-0.6715613007545471, -0.443665087223053], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004426.pt"}
{"idx": 4428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3jb5aI4ejo.mp4", "duration": 298.3814111111111, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where she is relaxing in a bubble bath.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where she is relaxing in a bubble bath.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 143.0], "base_se_norm": [-0.13533484935760498, -0.04149526357650757], "gt_se_raw": [130.0, 144.0], "gt_se_norm": [-0.12863200902938843, -0.034792423248291016], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004428.pt"}
{"idx": 4430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/khmmhvlCUNQ.mp4", "duration": 186.6, "prompt_text": "Please find the visual event described by the sentence 'When is the cursor pointing to the social media icons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cursor pointing to the social media icons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 39.0], "base_se_norm": [-0.6570203900337219, -0.581993579864502], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.5712754726409912, -0.5284030437469482], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004430.pt"}
{"idx": 4432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/khmmhvlCUNQ.mp4", "duration": 186.6, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'Future Students' navigation tab highlighted by the cursor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'Future Students' navigation tab highlighted by the cursor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.92497318983078], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [-0.24973207712173462, -0.22829586267471313], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004432.pt"}
{"idx": 4434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/khmmhvlCUNQ.mp4", "duration": 186.6, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the cursor highlights the site's search box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the cursor highlights the site's search box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 39.0], "base_se_norm": [-0.6570203900337219, -0.581993579864502], "gt_se_raw": [47.0, 53.0], "gt_se_norm": [-0.49624866247177124, -0.4319400191307068], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004434.pt"}
{"idx": 4436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KItqbZXlrdY.mp4", "duration": 99.84974444444444, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the ingredients are added to the mashed potatoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the ingredients are added to the mashed potatoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 27.0], "base_se_norm": [-0.679518461227417, -0.459187388420105], "gt_se_raw": [17.0, 27.0], "gt_se_norm": [-0.6594883799552917, -0.459187388420105], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004436.pt"}
{"idx": 4438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KItqbZXlrdY.mp4", "duration": 99.84974444444444, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the croquettes are being coated with flour, egg, and breadcrumbs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the croquettes are being coated with flour, egg, and breadcrumbs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 56.0], "base_se_norm": [-0.09864568710327148, 0.12168538570404053], "gt_se_raw": [45.0, 56.0], "gt_se_norm": [-0.09864568710327148, 0.12168538570404053], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004438.pt"}
{"idx": 4440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UlgYIQrcYnQ.mp4", "duration": 25.442083333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a person reaching into a jar to retrieve a pickle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person reaching into a jar to retrieve a pickle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.5283405184745789, -0.13529092073440552], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.4497305750846863, -0.2139008641242981], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004440.pt"}
{"idx": 4442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UlgYIQrcYnQ.mp4", "duration": 25.442083333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the Venus flytrap close its trap on the pickle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the Venus flytrap close its trap on the pickle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.056681036949157715, 0.25775861740112305], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [0.10053884983062744, 0.25775861740112305], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004442.pt"}
{"idx": 4444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UlgYIQrcYnQ.mp4", "duration": 25.442083333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a person is eating a pickle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a person is eating a pickle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.8427801728248596], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9213900566101074, -0.7641702890396118], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004444.pt"}
{"idx": 4446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0rrS51ry2s.mp4", "duration": 112.82103333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a person is shaking a maraca.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a person is shaking a maraca.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 36.0], "base_se_norm": [-0.468184232711792, -0.36182111501693726], "gt_se_raw": [30.0, 36.0], "gt_se_norm": [-0.468184232711792, -0.36182111501693726], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004446.pt"}
{"idx": 4448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0rrS51ry2s.mp4", "duration": 112.82103333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman feeds the baby with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman feeds the baby with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 60.0], "base_se_norm": [-0.09591323137283325, 0.06363153457641602], "gt_se_raw": [51.0, 59.0], "gt_se_norm": [-0.09591323137283325, 0.04590427875518799], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004448.pt"}
{"idx": 4450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BnZtKCcMxzE.mp4", "duration": 128.9288, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a close-up of the smart speakers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a close-up of the smart speakers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 41.0], "base_se_norm": [-0.7673134207725525, -0.36399006843566895], "gt_se_raw": [15.0, 41.0], "gt_se_norm": [-0.7673134207725525, -0.36399006843566895], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004450.pt"}
{"idx": 4452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BnZtKCcMxzE.mp4", "duration": 128.9288, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the smart speaker whispers a secret about the Property Brothers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the smart speaker whispers a secret about the Property Brothers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 41.0], "base_se_norm": [-0.7518010139465332, -0.36399006843566895], "gt_se_raw": [103.0, 113.0], "gt_se_norm": [0.5977810621261597, 0.7529054880142212], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004452.pt"}
{"idx": 4454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BnZtKCcMxzE.mp4", "duration": 128.9288, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Alexa tells a joke about seeing Twitch naked.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Alexa tells a joke about seeing Twitch naked.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 64.0], "base_se_norm": [-0.16232836246490479, -0.007203996181488037], "gt_se_raw": [117.0, 119.0], "gt_se_norm": [0.8149552345275879, 0.8459800481796265], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004454.pt"}
{"idx": 4456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_PklVas9cA.mp4", "duration": 141.161, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the three of them are lifting and carrying the couch up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the three of them are lifting and carrying the couch up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 114.0], "base_se_norm": [-0.22074788808822632, 0.6151770353317261], "gt_se_raw": [58.0, 80.0], "gt_se_norm": [-0.17824327945709229, 0.13345754146575928], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004456.pt"}
{"idx": 4458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_PklVas9cA.mp4", "duration": 141.161, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when Chandler gets angry and yells, 'Shut up, shut up, shut up!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when Chandler gets angry and yells, 'Shut up, shut up, shut up!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [-0.3199254274368286, -0.2632525563240051], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.33181262016296387, 0.3743172883987427], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004458.pt"}
{"idx": 4460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ixh8KqEr6LE.mp4", "duration": 115.03158333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two poles of the mop connected?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two poles of the mop connected?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.5305637121200562, -0.4610176086425781], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.860907793045044, -0.7565885782241821], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004460.pt"}
{"idx": 4462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ixh8KqEr6LE.mp4", "duration": 115.03158333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the green concentrate begin to infuse into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the green concentrate begin to infuse into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 74.0], "base_se_norm": [0.2170569896697998, 0.28660309314727783], "gt_se_raw": [70.0, 73.0], "gt_se_norm": [0.2170569896697998, 0.2692166566848755], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004462.pt"}
{"idx": 4464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ixh8KqEr6LE.mp4", "duration": 115.03158333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the person is mopping the wooden floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the person is mopping the wooden floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 96.0], "base_se_norm": [0.6169471740722656, 0.6691068410873413], "gt_se_raw": [93.0, 96.0], "gt_se_norm": [0.6169471740722656, 0.6691068410873413], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004464.pt"}
{"idx": 4466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HoKkasEyDOI.mp4", "duration": 195.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the phrase 'Lieseke also brought up the threat of gang activity' highlighted in the newspaper article?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the phrase 'Lieseke also brought up the threat of gang activity' highlighted in the newspaper article?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7448437213897705, -0.7142249345779419], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.7448437213897705, -0.7142249345779419], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004466.pt"}
{"idx": 4468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HoKkasEyDOI.mp4", "duration": 195.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a police officer violently pull a student out of their desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a police officer violently pull a student out of their desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [-0.3672124147415161, -0.3263874053955078], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [-0.3672124147415161, -0.3263874053955078], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004468.pt"}
{"idx": 4470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9MO1aY1xC80.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When can we see the ants placing their offerings on the stone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the ants placing their offerings on the stone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.6253746747970581, -0.550449550151825], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.6253746747970581, -0.5254745483398438], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004470.pt"}
{"idx": 4472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9MO1aY1xC80.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a single ant realizes he is lost and panics.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a single ant realizes he is lost and panics.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [0.42357635498046875, 0.5734264850616455], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.3256743550300598, -0.25074928998947144], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004472.pt"}
{"idx": 4474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9MO1aY1xC80.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the lost ant rejoins the line and thanks the leader.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the lost ant rejoins the line and thanks the leader.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 63.0], "base_se_norm": [0.32367634773254395, 0.5734264850616455], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [0.5234764814376831, 0.5984015464782715], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004474.pt"}
{"idx": 4476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/65WL8FfN-Io.mp4", "duration": 174.17399999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the man and woman are dancing together on the bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the man and woman are dancing together on the bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 54.0], "base_se_norm": [-0.5981030464172363, -0.37993037700653076], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [-0.5062408447265625, -0.4717925786972046], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004476.pt"}
{"idx": 4478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/65WL8FfN-Io.mp4", "duration": 174.17399999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see several women drumming and dancing simultaneously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see several women drumming and dancing simultaneously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 121.0], "base_se_norm": [0.3549668788909912, 0.38941526412963867], "gt_se_raw": [120.0, 121.0], "gt_se_norm": [0.37793242931365967, 0.38941526412963867], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004478.pt"}
{"idx": 4480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v_dbomeVN2c.mp4", "duration": 29.904874999999997, "prompt_text": "Please find the visual event described by the sentence 'When do we see a woman taking a photo in front of the Towson University logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see a woman taking a photo in front of the Towson University logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.7993638515472412, -0.6656063795089722], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.7993638515472412, -0.6656063795089722], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004480.pt"}
{"idx": 4482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v_dbomeVN2c.mp4", "duration": 29.904874999999997, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman in sunglasses takes a selfie next to a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman in sunglasses takes a selfie next to a bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [0.13693833351135254, 0.2706958055496216], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [0.13693833351135254, 0.2706958055496216], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004482.pt"}
{"idx": 4484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v_dbomeVN2c.mp4", "duration": 29.904874999999997, "prompt_text": "Please find the visual event described by the sentence 'Find the part where someone is shown sharing the AR experience on their phone with the #TUproud hashtag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where someone is shown sharing the AR experience on their phone with the #TUproud hashtag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [0.6050894260406494, 0.9394831657409668], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [0.6050894260406494, 0.9394831657409668], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004484.pt"}
{"idx": 4486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGnIEvFtSCY.mp4", "duration": 147.71423333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the instructor helps the student sign the word 'art'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the instructor helps the student sign the word 'art'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 113.0], "base_se_norm": [0.38104498386383057, 0.529981255531311], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.6344292759895325, -0.5938103199005127], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004486.pt"}
{"idx": 4488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGnIEvFtSCY.mp4", "duration": 147.71423333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the instructor helps sign for 'restaurant'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the instructor helps sign for 'restaurant'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 113.0], "base_se_norm": [0.3675053119659424, 0.529981255531311], "gt_se_raw": [77.0, 82.0], "gt_se_norm": [0.04255354404449463, 0.110251784324646], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004488.pt"}
{"idx": 4490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pYekAIt9wW8.mp4", "duration": 39.63333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the woman and the child singing the days of the week for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman and the child singing the days of the week for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 18.0], "base_se_norm": [-1.0, -0.09167373180389404], "gt_se_raw": [6.0, 20.0], "gt_se_norm": [-0.6972245573997498, 0.00925147533416748], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004490.pt"}
{"idx": 4492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pYekAIt9wW8.mp4", "duration": 39.63333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are their hands moving along the bottom row of the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are their hands moving along the bottom row of the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 38.0], "base_se_norm": [0.21110177040100098, 0.9175777435302734], "gt_se_raw": [26.0, 38.0], "gt_se_norm": [0.31202685832977295, 0.9175777435302734], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004492.pt"}
{"idx": 4494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pYekAIt9wW8.mp4", "duration": 39.63333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in green guide the child's hand across the top row of the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in green guide the child's hand across the top row of the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 18.0], "base_se_norm": [-0.949537456035614, -0.09167373180389404], "gt_se_raw": [3.0, 18.0], "gt_se_norm": [-0.8486123085021973, -0.09167373180389404], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004494.pt"}
{"idx": 4496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dSf32CWJYBs.mp4", "duration": 110.735625, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a couple greets the people on the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a couple greets the people on the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 38.0], "base_se_norm": [-0.5123520493507385, -0.31368064880371094], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.45816898345947266, -0.403985857963562], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004496.pt"}
{"idx": 4498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dSf32CWJYBs.mp4", "duration": 110.735625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man on the bench say he is an atheist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man on the bench say he is an atheist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 55.0], "base_se_norm": [-0.042765140533447266, -0.006643056869506836], "gt_se_raw": [54.0, 55.0], "gt_se_norm": [-0.02470409870147705, -0.006643056869506836], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004498.pt"}
{"idx": 4500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A1T3dAvFeio.mp4", "duration": 165.0, "prompt_text": "Please find the visual event described by the sentence 'When do we see the pigs watching a movie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see the pigs watching a movie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 58.0], "base_se_norm": [-0.478787899017334, -0.2969697117805481], "gt_se_raw": [45.0, 59.0], "gt_se_norm": [-0.45454543828964233, -0.28484851121902466], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004500.pt"}
{"idx": 4502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A1T3dAvFeio.mp4", "duration": 165.0, "prompt_text": "Please find the visual event described by the sentence 'When does the red bird produce an egg to swap with the cabbage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the red bird produce an egg to swap with the cabbage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 126.0], "base_se_norm": [0.5030303001403809, 0.5272727012634277], "gt_se_raw": [124.0, 125.0], "gt_se_norm": [0.5030303001403809, 0.5151515007019043], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004502.pt"}
{"idx": 4504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A1T3dAvFeio.mp4", "duration": 165.0, "prompt_text": "Please find the visual event described by the sentence 'Show the birds swinging violently inside the cage, causing it to hit a chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the birds swinging violently inside the cage, causing it to hit a chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 106.0], "base_se_norm": [0.21212124824523926, 0.2848484516143799], "gt_se_raw": [100.0, 105.0], "gt_se_norm": [0.21212124824523926, 0.27272725105285645], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004504.pt"}
{"idx": 4506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebqxuyvp1qk.mp4", "duration": 39.005633333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl hold the bear and kiss its nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl hold the bear and kiss its nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [0.3844153881072998, 0.5382393598556519], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.4356900453567505, 0.5382393598556519], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004506.pt"}
{"idx": 4508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebqxuyvp1qk.mp4", "duration": 39.005633333333336, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the woman points to the cover of the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the woman points to the cover of the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 38.0], "base_se_norm": [0.6407885551452637, 0.9484364986419678], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [0.6407885551452637, 0.743337869644165], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004508.pt"}
{"idx": 4510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JnR2dpLnS14.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a climber scaling a sheer wall of ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a climber scaling a sheer wall of ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7180087566375732, -0.6841697692871094], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.7067291140556335, -0.6841697692871094], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004510.pt"}
{"idx": 4512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JnR2dpLnS14.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do two climbers cross a ladder bridge together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two climbers cross a ladder bridge together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 99.0], "base_se_norm": [0.02644813060760498, 0.11668527126312256], "gt_se_raw": [90.0, 98.0], "gt_se_norm": [0.015168428421020508, 0.10540568828582764], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004512.pt"}
{"idx": 4514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JnR2dpLnS14.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when climbers are wearing yellow oxygen masks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when climbers are wearing yellow oxygen masks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 153.0], "base_se_norm": [0.5904306173324585, 0.7257864475250244], "gt_se_raw": [150.0, 153.0], "gt_se_norm": [0.6919474601745605, 0.7257864475250244], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004514.pt"}
{"idx": 4516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J0HiXwK5s2k.mp4", "duration": 63.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the white lab coat begin to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the white lab coat begin to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 12.0], "base_se_norm": [-0.8730158805847168, -0.6190476417541504], "gt_se_raw": [3.0, 12.0], "gt_se_norm": [-0.9047619104385376, -0.6190476417541504], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004516.pt"}
{"idx": 4518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J0HiXwK5s2k.mp4", "duration": 63.0, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the woman in the pink vest conducting an experiment in a sterile hood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the woman in the pink vest conducting an experiment in a sterile hood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.238095223903656, -0.0476190447807312], "gt_se_raw": [23.0, 30.0], "gt_se_norm": [-0.2698412537574768, -0.0476190447807312], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004518.pt"}
{"idx": 4520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsSq3xiTWC8.mp4", "duration": 87.4874, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put the red bowl aside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put the red bowl aside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.4742099642753601, -0.40562868118286133], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.4742099642753601, -0.428489089012146], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004520.pt"}
{"idx": 4522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsSq3xiTWC8.mp4", "duration": 87.4874, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman put her finger to the girl's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman put her finger to the girl's mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 72.0], "base_se_norm": [0.5773700475692749, 0.6459513902664185], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [0.5773700475692749, 0.6459513902664185], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004522.pt"}
{"idx": 4524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsSq3xiTWC8.mp4", "duration": 87.4874, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put the dough into the metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put the dough into the metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 86.0], "base_se_norm": [-0.3827682137489319, 0.9659974575042725], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.8516952991485596, 0.9202766418457031], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004524.pt"}
{"idx": 4526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I2yW_VONWco.mp4", "duration": 406.0, "prompt_text": "Please find the visual event described by the sentence 'Show the process of kneading the dough with hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of kneading the dough with hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 82.0], "base_se_norm": [-0.6403940916061401, -0.5960590839385986], "gt_se_raw": [75.0, 82.0], "gt_se_norm": [-0.6305418610572815, -0.5960590839385986], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004526.pt"}
{"idx": 4528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I2yW_VONWco.mp4", "duration": 406.0, "prompt_text": "Please find the visual event described by the sentence 'Show when the baker is stacking the different colored dough layers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show when the baker is stacking the different colored dough layers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 201.0], "base_se_norm": [-0.10344827175140381, -0.009852230548858643], "gt_se_raw": [186.0, 200.0], "gt_se_norm": [-0.0837438702583313, -0.014778316020965576], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004528.pt"}
{"idx": 4530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aJV1WVWbyNU.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the two cats exiting the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the two cats exiting the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6954494714736938, -0.66161048412323], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.7067291140556335, -0.6503309011459351], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004530.pt"}
{"idx": 4532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aJV1WVWbyNU.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see one cat playfully pouncing on the other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see one cat playfully pouncing on the other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 91.0], "base_se_norm": [-0.018670499324798584, 0.02644813060760498], "gt_se_raw": [88.0, 92.0], "gt_se_norm": [-0.007390797138214111, 0.03772783279418945], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004532.pt"}
{"idx": 4534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aJV1WVWbyNU.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the two cats seen walking together in the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two cats seen walking together in the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 88.0], "base_se_norm": [-0.6052122712135315, -0.007390797138214111], "gt_se_raw": [73.0, 80.0], "gt_se_norm": [-0.17658555507659912, -0.09762799739837646], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004534.pt"}
{"idx": 4536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6lZBoR5gvs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment that quantifies the employment gap by stating one group's rate is 'nearly half' of the other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment that quantifies the employment gap by stating one group's rate is 'nearly half' of the other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.03166300058364868, 0.355671763420105], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.03166300058364868, 0.355671763420105], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004536.pt"}
{"idx": 4538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6lZBoR5gvs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the speaker explicitly refers to 'Americans without disabilities'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the speaker explicitly refers to 'Americans without disabilities'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.03166300058364868, 0.25883805751800537], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9031662940979004, -0.7094988822937012], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004538.pt"}
{"idx": 4540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S1Cn8vIGlLo.mp4", "duration": 66.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is the woman singing 'If You're Happy and You Know It'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is the woman singing 'If You're Happy and You Know It'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 66.0], "base_se_norm": [-0.9700748324394226, 0.9750622510910034], "gt_se_raw": [46.0, 62.0], "gt_se_norm": [0.3765585422515869, 0.8553615808486938], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004540.pt"}
{"idx": 4542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S1Cn8vIGlLo.mp4", "duration": 66.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman try to get the baby to clap by repeatedly saying 'Eli' and clapping her own hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman try to get the baby to clap by repeatedly saying 'Eli' and clapping her own hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 12.0], "base_se_norm": [-0.9401496052742004, -0.6408977508544922], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.910224437713623, -0.7605985403060913], "start_token_span": [85, 88], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004542.pt"}
{"idx": 4544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S1Cn8vIGlLo.mp4", "duration": 66.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the baby smiles in response to the woman's interaction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the baby smiles in response to the woman's interaction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.49127185344696045, -0.31172072887420654], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.4014962911605835, -0.34164589643478394], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004544.pt"}
{"idx": 4546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9kxmWv1AV5I.mp4", "duration": 62.10370833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when chopped garlic is added to the frying chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when chopped garlic is added to the frying chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.6135496497154236, -0.5169370770454407], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.6135496497154236, -0.5169370770454407], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004546.pt"}
{"idx": 4548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9kxmWv1AV5I.mp4", "duration": 62.10370833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the grated parmesan is being mixed into the pasta.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the grated parmesan is being mixed into the pasta.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 44.0], "base_se_norm": [0.19155526161193848, 0.41698455810546875], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [0.19155526161193848, 0.3847804069519043], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004548.pt"}
{"idx": 4550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CYgs0ccffVw.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the girl use sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl use sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 16.0], "base_se_norm": [-0.5544553995132446, 0.1881188154220581], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.5544553995132446, -0.10891085863113403], "start_token_span": [68, 71], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004550.pt"}
{"idx": 4552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CYgs0ccffVw.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the girl press a button to change the letter on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl press a button to change the letter on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.41089117527008057, 0.6336634159088135], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [0.11386144161224365, 0.2623763084411621], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004552.pt"}
{"idx": 4554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t3NgDZTVPR0.mp4", "duration": 133.84204166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the boy using a yellow walker on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the boy using a yellow walker on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.9402280449867249, -0.8655130863189697], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.9402280449867249, -0.8804560899734497], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004554.pt"}
{"idx": 4556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t3NgDZTVPR0.mp4", "duration": 133.84204166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is there a static shot of the modified green toy tractor parked on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there a static shot of the modified green toy tractor parked on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.5815963745117188, -0.5068813562393188], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.5815963745117188, -0.5068813562393188], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004556.pt"}
{"idx": 4558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t3NgDZTVPR0.mp4", "duration": 133.84204166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy is driving the toy tractor as a crowd watches and claps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy is driving the toy tractor as a crowd watches and claps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 128.0], "base_se_norm": [0.6736146211624146, 0.9127024412155151], "gt_se_raw": [112.0, 126.0], "gt_se_norm": [0.6736146211624146, 0.8828164339065552], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004558.pt"}
{"idx": 4560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBLCi5Tdpbw.mp4", "duration": 138.24, "prompt_text": "Please find the visual event described by the sentence 'Show the process of folding in all four corners towards the center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of folding in all four corners towards the center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 41.0], "base_se_norm": [-0.6527777910232544, -0.40682870149612427], "gt_se_raw": [24.0, 44.0], "gt_se_norm": [-0.6527777910232544, -0.36342597007751465], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004560.pt"}
{"idx": 4562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBLCi5Tdpbw.mp4", "duration": 138.24, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the corners are bent to create the alligator's eyes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the corners are bent to create the alligator's eyes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 105.0], "base_se_norm": [0.14293980598449707, 0.5190972089767456], "gt_se_raw": [79.0, 87.0], "gt_se_norm": [0.14293980598449707, 0.25868046283721924], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004562.pt"}
{"idx": 4564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D8Laqglo1ig.mp4", "duration": 143.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'When are two women sitting on the floor comforting each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women sitting on the floor comforting each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.5272305607795715, -0.4716106653213501], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.5272305607795715, -0.4716106653213501], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004564.pt"}
{"idx": 4566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D8Laqglo1ig.mp4", "duration": 143.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene showing a bar chart about disabled parking fines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene showing a bar chart about disabled parking fines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 103.0], "base_se_norm": [0.27925848960876465, 0.43221330642700195], "gt_se_raw": [91.0, 103.0], "gt_se_norm": [0.2653534412384033, 0.43221330642700195], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004566.pt"}
{"idx": 4568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D8Laqglo1ig.mp4", "duration": 143.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a white car pulls into a disabled parking spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a white car pulls into a disabled parking spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8192352056503296, -0.7775202989578247], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.8192352056503296, -0.7914252281188965], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004568.pt"}
{"idx": 4570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6cbRfTIrmc.mp4", "duration": 23.095325726904672, "prompt_text": "Please find the visual event described by the sentence 'At what time can a person be heard sneezing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time can a person be heard sneezing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.38556182384490967, 0.558756947517395], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.38556182384490967, 0.558756947517395], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004570.pt"}
{"idx": 4572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6cbRfTIrmc.mp4", "duration": 23.095325726904672, "prompt_text": "Please find the visual event described by the sentence 'When is the camera zooming in on the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the camera zooming in on the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 16.0], "base_se_norm": [-0.7402071952819824, 0.38556182384490967], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.7402071952819824, -0.5670119524002075], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004572.pt"}
{"idx": 4574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PGKmexNTHNE.mp4", "duration": 12.24, "prompt_text": "Please find the visual event described by the sentence 'When does the lamp jump on and squash the letter 'I'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the lamp jump on and squash the letter 'I'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.673202633857727, -0.3464052081108093], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.5098038911819458, -0.3464052081108093], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004574.pt"}
{"idx": 4576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PGKmexNTHNE.mp4", "duration": 12.24, "prompt_text": "Please find the visual event described by the sentence 'Show the sequence where the lamp hops from off-screen to the left of the letter 'X'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the sequence where the lamp hops from off-screen to the left of the letter 'X'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.8366013169288635, 0.30718958377838135], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.8366013169288635, -0.5098038911819458], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004576.pt"}
{"idx": 4578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PGKmexNTHNE.mp4", "duration": 12.24, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'Animation Studios' fade in?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'Animation Studios' fade in?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [0.30718958377838135, 0.6339869499206543], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [0.47058820724487305, 0.6339869499206543], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004578.pt"}
{"idx": 4580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5Ddap2Pyhtw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment one man comforts the other by resting his head on his shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment one man comforts the other by resting his head on his shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 43.0], "base_se_norm": [-0.6659687161445618, -0.46802425384521484], "gt_se_raw": [39.0, 48.0], "gt_se_norm": [-0.5175103545188904, -0.40616661310195923], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004580.pt"}
{"idx": 4582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5Ddap2Pyhtw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man is shot in the back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man is shot in the back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 86.0], "base_se_norm": [0.0020939111709594727, 0.06395149230957031], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.051579952239990234, 0.07632303237915039], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004582.pt"}
{"idx": 4584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LzKwClRvRco.mp4", "duration": 291.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a convoy of police trucks driving on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a convoy of police trucks driving on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.9793908596038818, -0.9106938242912292], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.972521185874939, -0.9244332313537598], "start_token_span": [74, 77], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004584.pt"}
{"idx": 4586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LzKwClRvRco.mp4", "duration": 291.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the aerial shot of the congested highway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the aerial shot of the congested highway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 56.0], "base_se_norm": [-0.7183420658111572, -0.6152965426445007], "gt_se_raw": [41.0, 48.0], "gt_se_norm": [-0.7183420658111572, -0.6702541708946228], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004586.pt"}
{"idx": 4588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LzKwClRvRco.mp4", "duration": 291.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the gunfight erupt on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 184.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the gunfight erupt on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 184.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 184.0], "base_se_norm": [-0.1893748641014099, 0.2640256881713867], "gt_se_raw": [225.0, 231.0], "gt_se_norm": [0.5456835031509399, 0.5869017839431763], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004588.pt"}
{"idx": 4590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TIQZjO5dfnQ.mp4", "duration": 74.0, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where Iron Man flies through the city.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where Iron Man flies through the city.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [-0.35135138034820557, -0.29729729890823364], "gt_se_raw": [25.0, 26.0], "gt_se_norm": [-0.3243243098258972, -0.29729729890823364], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004590.pt"}
{"idx": 4592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TIQZjO5dfnQ.mp4", "duration": 74.0, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a man with a beard stares intensely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a man with a beard stares intensely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 48.0], "base_se_norm": [0.24324321746826172, 0.29729723930358887], "gt_se_raw": [46.0, 47.0], "gt_se_norm": [0.24324321746826172, 0.2702702283859253], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004592.pt"}
{"idx": 4594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MmVXLM97YL0.mp4", "duration": 106.6065, "prompt_text": "Please find the visual event described by the sentence 'When is the hazelnut spread added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the hazelnut spread added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.8311547636985779, -0.7936335802078247], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.8311547636985779, -0.7936335802078247], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004594.pt"}
{"idx": 4596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MmVXLM97YL0.mp4", "duration": 106.6065, "prompt_text": "Please find the visual event described by the sentence 'At what point are the crushed sandwich cookies poured into the mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the crushed sandwich cookies poured into the mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.08073145151138306, -0.04321032762527466], "gt_se_raw": [49.0, 50.0], "gt_se_norm": [-0.08073145151138306, -0.06197088956832886], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004596.pt"}
{"idx": 4598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MmVXLM97YL0.mp4", "duration": 106.6065, "prompt_text": "Please find the visual event described by the sentence 'When are the sprinkles added to the birthday cake batter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the sprinkles added to the birthday cake batter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 75.0], "base_se_norm": [0.35076189041137695, 0.4070436954498291], "gt_se_raw": [73.0, 74.0], "gt_se_norm": [0.3695225715637207, 0.3882831335067749], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004598.pt"}
{"idx": 4600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/avY2zI8fhUc.mp4", "duration": 144.24, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man in a wheelchair wearing a black shirt as he rolls through the office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man in a wheelchair wearing a black shirt as he rolls through the office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 34.0], "base_se_norm": [-0.8336106538772583, -0.5285634994506836], "gt_se_raw": [21.0, 34.0], "gt_se_norm": [-0.7088186740875244, -0.5285634994506836], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004600.pt"}
{"idx": 4602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/avY2zI8fhUc.mp4", "duration": 144.24, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the 'Upcoming Events' page on a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the 'Upcoming Events' page on a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 68.0], "base_se_norm": [-0.20965063571929932, -0.05712705850601196], "gt_se_raw": [57.0, 69.0], "gt_se_norm": [-0.20965063571929932, -0.04326128959655762], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004602.pt"}
{"idx": 4604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZfjrd-B0ho.mp4", "duration": 163.99716666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe someone hitting a ball during baseball practice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe someone hitting a ball during baseball practice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.8292653560638428, -0.8048746585845947], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [-0.20730334520339966, -0.1829127073287964], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004604.pt"}
{"idx": 4606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZfjrd-B0ho.mp4", "duration": 163.99716666666666, "prompt_text": "Please find the visual event described by the sentence 'When is a person in a pink shirt speaking into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person in a pink shirt speaking into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 61.0], "base_se_norm": [-0.3170613646507263, -0.256084680557251], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.036603331565856934, 0.12197065353393555], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004606.pt"}
{"idx": 4608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZfjrd-B0ho.mp4", "duration": 163.99716666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a group of people start throwing colored powder in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a group of people start throwing colored powder in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 111.0], "base_se_norm": [0.2927052974700928, 0.35368192195892334], "gt_se_raw": [106.0, 111.0], "gt_se_norm": [0.2927052974700928, 0.35368192195892334], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004608.pt"}
{"idx": 4610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hWskc5_mPSI.mp4", "duration": 14.948384501943961, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where someone draws the letter 'A' in the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where someone draws the letter 'A' in the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.19723767042160034, 0.33793723583221436], "gt_se_raw": [6.0, 10.0], "gt_se_norm": [-0.19723767042160034, 0.33793723583221436], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004610.pt"}
{"idx": 4612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hWskc5_mPSI.mp4", "duration": 14.948384501943961, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a person running along the shoreline while inside a giant yellow 'A'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a person running along the shoreline while inside a giant yellow 'A'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.3310313820838928], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.5986188650131226, -0.3310313820838928], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004612.pt"}
{"idx": 4614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bCO3S2_s07w.mp4", "duration": 176.20936666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy using the device to communicate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy using the device to communicate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 117.0], "base_se_norm": [-0.035238564014434814, 0.32796573638916016], "gt_se_raw": [3.0, 18.0], "gt_se_norm": [-0.9659495949745178, -0.7956975698471069], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004614.pt"}
{"idx": 4616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bCO3S2_s07w.mp4", "duration": 176.20936666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the boy is leaning his head on his arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the boy is leaning his head on his arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 138.0], "base_se_norm": [0.32796573638916016, 0.5663185119628906], "gt_se_raw": [150.0, 158.0], "gt_se_norm": [0.7025201320648193, 0.7933212518692017], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004616.pt"}
{"idx": 4618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bCO3S2_s07w.mp4", "duration": 176.20936666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the boy lean back and smile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy lean back and smile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 146.0], "base_se_norm": [0.6230692863464355, 0.657119631767273], "gt_se_raw": [102.0, 105.0], "gt_se_norm": [0.15771377086639404, 0.19176411628723145], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004618.pt"}
{"idx": 4620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CGAB2warrpA.mp4", "duration": 120.98753333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where two boys are walking together on a hilltop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where two boys are walking together on a hilltop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 44.0], "base_se_norm": [-0.3883667588233948, -0.2726523280143738], "gt_se_raw": [38.0, 44.0], "gt_se_norm": [-0.3718361258506775, -0.2726523280143738], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004620.pt"}
{"idx": 4622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CGAB2warrpA.mp4", "duration": 120.98753333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the scene showing the stairs to heaven appearing in the clouds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the scene showing the stairs to heaven appearing in the clouds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 60.0], "base_se_norm": [-0.07428479194641113, -0.008162260055541992], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [-0.07428479194641113, -0.008162260055541992], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004622.pt"}
{"idx": 4624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iFIsvjt9NSs.mp4", "duration": 176.00916666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the baby playing with the blue doll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the baby playing with the blue doll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 72.0], "base_se_norm": [-0.7272869348526001, -0.1818608045578003], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.7272869348526001, -0.6363825798034668], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004624.pt"}
{"idx": 4626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iFIsvjt9NSs.mp4", "duration": 176.00916666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the baby is holding the green toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the baby is holding the green toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 18.0], "base_se_norm": [-0.9431847929954529, -0.7954652309417725], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.556841254234314, -0.534115195274353], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004626.pt"}
{"idx": 4628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iFIsvjt9NSs.mp4", "duration": 176.00916666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the baby picks up the blue ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the baby picks up the blue ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 126.0], "base_se_norm": [0.26129794120788574, 0.4317436218261719], "gt_se_raw": [124.0, 127.0], "gt_se_norm": [0.4090174436569214, 0.44310665130615234], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004628.pt"}
{"idx": 4630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uqxzi_ghjgc.mp4", "duration": 110.0, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl is unhappy about the vegetables on her plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl is unhappy about the vegetables on her plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.3999999761581421, -0.3272727131843567], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.418181836605072, -0.3272727131843567], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004630.pt"}
{"idx": 4632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uqxzi_ghjgc.mp4", "duration": 110.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the girl is happily playing with building blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the girl is happily playing with building blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 63.0], "base_se_norm": [0.018181800842285156, 0.1454545259475708], "gt_se_raw": [55.0, 60.0], "gt_se_norm": [0.0, 0.09090912342071533], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004632.pt"}
{"idx": 4634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uYOpUXb8tYU.mp4", "duration": 184.4843, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is holding up a book titled 'Planet Earth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is holding up a book titled 'Planet Earth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9674769043922424, -0.9349538087844849], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9566358923912048, -0.9349538087844849], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004634.pt"}
{"idx": 4636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uYOpUXb8tYU.mp4", "duration": 184.4843, "prompt_text": "Please find the visual event described by the sentence 'Show the sequence where the balloon rocket is being prepared.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the sequence where the balloon rocket is being prepared.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 156.0], "base_se_norm": [0.45269811153411865, 0.691200852394104], "gt_se_raw": [127.0, 155.0], "gt_se_norm": [0.3768109083175659, 0.6803598403930664], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004636.pt"}
{"idx": 4638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uYOpUXb8tYU.mp4", "duration": 184.4843, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man is walking through a large, circular museum lobby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man is walking through a large, circular museum lobby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.7506563067436218, -0.7072921991348267], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7398152351379395, -0.7072921991348267], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004638.pt"}
{"idx": 4640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-l0tT9PVadQ.mp4", "duration": 126.29283333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person is making a series of parallel cuts into the side of the folded blue paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person is making a series of parallel cuts into the side of the folded blue paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [-0.287370502948761, -0.14484459161758423], "gt_se_raw": [46.0, 54.0], "gt_se_norm": [-0.271534264087677, -0.14484459161758423], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004640.pt"}
{"idx": 4642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-l0tT9PVadQ.mp4", "duration": 126.29283333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person use a black marker to draw details on the yellow paper cutouts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person use a black marker to draw details on the yellow paper cutouts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 108.0], "base_se_norm": [0.6311297416687012, 0.7103108167648315], "gt_se_raw": [103.0, 107.0], "gt_se_norm": [0.6311297416687012, 0.6944746971130371], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004642.pt"}
{"idx": 4644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4h-PtwiqWE.mp4", "duration": 9.8, "prompt_text": "Please find the visual event described by the sentence 'When does the girl touch the white shelf with her right hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl touch the white shelf with her right hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.5918367505073547, -0.18367350101470947], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.5918367505073547, -0.3877550959587097], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004644.pt"}
{"idx": 4646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4h-PtwiqWE.mp4", "duration": 9.8, "prompt_text": "Please find the visual event described by the sentence 'At what time does the girl walk towards the left side of the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the girl walk towards the left side of the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [0.42857134342193604, 0.8367346525192261], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [0.42857134342193604, 0.8367346525192261], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004646.pt"}
{"idx": 4648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4h-PtwiqWE.mp4", "duration": 9.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the girl adjusts a white card near the top of the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the girl adjusts a white card near the top of the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.3877550959587097], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.795918345451355, -0.3877550959587097], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004648.pt"}
{"idx": 4650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OINa46HeWg8.mp4", "duration": 130.04657777777777, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where four people are having a meal at an outdoor restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where four people are having a meal at an outdoor restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 43.0], "base_se_norm": [-0.5847641825675964, -0.33869850635528564], "gt_se_raw": [26.0, 43.0], "gt_se_norm": [-0.600143313407898, -0.33869850635528564], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004650.pt"}
{"idx": 4652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OINa46HeWg8.mp4", "duration": 130.04657777777777, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a woman in a skull-patterned dress bowls a strike?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a woman in a skull-patterned dress bowls a strike?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 78.0], "base_se_norm": [0.09191644191741943, 0.1995701789855957], "gt_se_raw": [71.0, 77.0], "gt_se_norm": [0.09191644191741943, 0.18419098854064941], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004652.pt"}
{"idx": 4654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTMKUe8VF0M.mp4", "duration": 401.4343666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the red and white striped shirt pick up the picture card and hold it up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the red and white striped shirt pick up the picture card and hold it up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.8854109048843384, -0.8654823303222656], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.8953751921653748, -0.8704645037651062], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004654.pt"}
{"idx": 4656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTMKUe8VF0M.mp4", "duration": 401.4343666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the boy standing up and holding a white toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy standing up and holding a white toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 118.0], "base_se_norm": [-0.496804416179657, -0.4121081233024597], "gt_se_raw": [118.0, 124.0], "gt_se_norm": [-0.4121081233024597, -0.38221532106399536], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004656.pt"}
{"idx": 4658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTMKUe8VF0M.mp4", "duration": 401.4343666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the camouflage shirt take the yellow-bordered card from the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the camouflage shirt take the yellow-bordered card from the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 37.0], "base_se_norm": [-0.855518102645874, -0.8156610131263733], "gt_se_raw": [65.0, 66.0], "gt_se_norm": [-0.6761612892150879, -0.6711791157722473], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004658.pt"}
{"idx": 4660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zE0qp3Yje6Q.mp4", "duration": 170.16, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where multiple adult turtles are resting on the sandy bottom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where multiple adult turtles are resting on the sandy bottom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 47.0], "base_se_norm": [-0.5768688321113586, -0.4475787878036499], "gt_se_raw": [36.0, 42.0], "gt_se_norm": [-0.5768688321113586, -0.5063469409942627], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004660.pt"}
{"idx": 4662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zE0qp3Yje6Q.mp4", "duration": 170.16, "prompt_text": "Please find the visual event described by the sentence 'Find the wide shot of many baby turtles crawling on the sand towards the sea at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the wide shot of many baby turtles crawling on the sand towards the sea at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 89.0], "base_se_norm": [-0.05970853567123413, 0.0460742712020874], "gt_se_raw": [80.0, 84.0], "gt_se_norm": [-0.05970853567123413, -0.012693941593170166], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004662.pt"}
{"idx": 4664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgfTgj_muvw.mp4", "duration": 130.4637137137137, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the statistic that people spend 40 hours a week on their phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the statistic that people spend 40 hours a week on their phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.8926904797554016, -0.7700510025024414], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.8926904797554016, -0.7700510025024414], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004664.pt"}
{"idx": 4666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgfTgj_muvw.mp4", "duration": 130.4637137137137, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a man in a wheelchair is unable to reach the sink in a bathroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a man in a wheelchair is unable to reach the sink in a bathroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 65.0], "base_se_norm": [-0.26416319608688354, -0.0035543441772460938], "gt_se_raw": [48.0, 65.0], "gt_se_norm": [-0.26416319608688354, -0.0035543441772460938], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004666.pt"}
{"idx": 4668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgfTgj_muvw.mp4", "duration": 130.4637137137137, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a red brushstroke covers a group of people with different disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a red brushstroke covers a group of people with different disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 101.0], "base_se_norm": [0.4563436508178711, 0.5483232736587524], "gt_se_raw": [94.0, 101.0], "gt_se_norm": [0.4410136938095093, 0.5483232736587524], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004668.pt"}
{"idx": 4670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ND63UPelkIw.mp4", "duration": 143.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the dog is pulling a man in a wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the dog is pulling a man in a wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 80.0], "base_se_norm": [-0.0829087495803833, 0.1116257905960083], "gt_se_raw": [79.0, 91.0], "gt_se_norm": [0.09773039817810059, 0.2644742727279663], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004670.pt"}
{"idx": 4672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ND63UPelkIw.mp4", "duration": 143.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the dog hops around on its hind legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the dog hops around on its hind legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 118.0], "base_se_norm": [0.43121814727783203, 0.6396479606628418], "gt_se_raw": [105.0, 117.0], "gt_se_norm": [0.4590088129043579, 0.6257526874542236], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004672.pt"}
{"idx": 4674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EIYDPwY3gNk.mp4", "duration": 85.96, "prompt_text": "Please find the visual event described by the sentence 'When is a woman watching TV alone while covering herself with a blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman watching TV alone while covering herself with a blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.813866913318634], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.8604001998901367], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004674.pt"}
{"idx": 4676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EIYDPwY3gNk.mp4", "duration": 85.96, "prompt_text": "Please find the visual event described by the sentence 'When does the man bring over a green stool to sit next to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man bring over a green stool to sit next to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.1391344666481018, -0.06933456659317017], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.1391344666481018, -0.09260118007659912], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004676.pt"}
{"idx": 4678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EIYDPwY3gNk.mp4", "duration": 85.96, "prompt_text": "Please find the visual event described by the sentence 'When does the woman get scared by the movie and hug the man for comfort?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman get scared by the movie and hug the man for comfort?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 83.0], "base_se_norm": [0.5123313665390015, 0.9311307668685913], "gt_se_raw": [66.0, 83.0], "gt_se_norm": [0.5355979204177856, 0.9311307668685913], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004678.pt"}
{"idx": 4680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gIUSfQ34Hj4.mp4", "duration": 161.3612, "prompt_text": "Please find the visual event described by the sentence 'When is the shaving foam added to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the shaving foam added to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 30.0], "base_se_norm": [-0.7149252891540527, -0.6281634569168091], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [-0.7149252891540527, -0.6405580043792725], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004680.pt"}
{"idx": 4682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gIUSfQ34Hj4.mp4", "duration": 161.3612, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person starts kneading the mixture with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person starts kneading the mixture with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 96.0], "base_se_norm": [0.07832610607147217, 0.18987703323364258], "gt_se_raw": [86.0, 93.0], "gt_se_norm": [0.06593155860900879, 0.15269339084625244], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004682.pt"}
{"idx": 4684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFJAuAWxuvI.mp4", "duration": 178.57840000000002, "prompt_text": "Please find the visual event described by the sentence 'When is brown sugar added to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is brown sugar added to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [-0.6752126812934875, -0.6416139602661133], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.664013147354126, -0.6416139602661133], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004684.pt"}
{"idx": 4686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFJAuAWxuvI.mp4", "duration": 178.57840000000002, "prompt_text": "Please find the visual event described by the sentence 'When are the chocolate chips being folded into the dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the chocolate chips being folded into the dough?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [0.11995625495910645, 0.25435101985931396], "gt_se_raw": [105.0, 112.0], "gt_se_norm": [0.17595410346984863, 0.25435101985931396], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004686.pt"}
{"idx": 4688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFJAuAWxuvI.mp4", "duration": 178.57840000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cookies are being baked in the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cookies are being baked in the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 148.0], "base_se_norm": [0.5007413625717163, 0.6575353145599365], "gt_se_raw": [138.0, 147.0], "gt_se_norm": [0.5455396175384521, 0.6463357210159302], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004688.pt"}
{"idx": 4690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5lgHJB1lwYs.mp4", "duration": 68.36828495161828, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a woman is doing jumping jacks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a woman is doing jumping jacks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.35642677545547485, -0.26866674423217773], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.35642677545547485, -0.26866674423217773], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004690.pt"}
{"idx": 4692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5lgHJB1lwYs.mp4", "duration": 68.36828495161828, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a woman in a purple top is doing squats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a woman in a purple top is doing squats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [0.1116265058517456, 0.2578932046890259], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [0.19938647747039795, 0.2578932046890259], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004692.pt"}
{"idx": 4694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UhQ9HquDNEM.mp4", "duration": 137.08, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man coming home and closing the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man coming home and closing the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.8978698253631592, -0.781149685382843], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.8978698253631592, -0.8249197602272034], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004694.pt"}
{"idx": 4696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UhQ9HquDNEM.mp4", "duration": 137.08, "prompt_text": "Please find the visual event described by the sentence 'When does the cat start rolling around on the floor with the catnip pouch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat start rolling around on the floor with the catnip pouch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 67.0], "base_se_norm": [-0.34344911575317383, -0.022468626499176025], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [-0.34344911575317383, -0.241318941116333], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004696.pt"}
{"idx": 4698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UhQ9HquDNEM.mp4", "duration": 137.08, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cat, in its excitement, knocks over the Christmas tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cat, in its excitement, knocks over the Christmas tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 123.0], "base_se_norm": [0.707032322883606, 0.7945724725723267], "gt_se_raw": [120.0, 123.0], "gt_se_norm": [0.7508023977279663, 0.7945724725723267], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004698.pt"}
{"idx": 4700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-u_KZ1BCaTw.mp4", "duration": 160.52703333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the text 'Implicit discrimination' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the text 'Implicit discrimination' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.5141005516052246, -0.4642646908760071], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.5141005516052246, -0.47672367095947266], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004700.pt"}
{"idx": 4702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-u_KZ1BCaTw.mp4", "duration": 160.52703333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the footage of the yellow train with the Sydney Harbour Bridge in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the footage of the yellow train with the Sydney Harbour Bridge in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 92.0], "base_se_norm": [-0.02820110321044922, 0.14622437953948975], "gt_se_raw": [78.0, 92.0], "gt_se_norm": [-0.02820110321044922, 0.14622437953948975], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004702.pt"}
{"idx": 4704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebyf6FzKEiI.mp4", "duration": 115.2151, "prompt_text": "Please find the visual event described by the sentence 'When is the woman singing 'You are my sunshine' to the baby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman singing 'You are my sunshine' to the baby?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, -0.37508195638656616], "gt_se_raw": [9.0, 30.0], "gt_se_norm": [-0.8437705039978027, -0.4792349338531494], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004704.pt"}
{"idx": 4706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebyf6FzKEiI.mp4", "duration": 115.2151, "prompt_text": "Please find the visual event described by the sentence 'Show the woman in black gently tapping on the child's chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman in black gently tapping on the child's chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 35.0], "base_se_norm": [-1.0, -0.3924407362937927], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.8437705039978027, -0.7396174669265747], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004706.pt"}
{"idx": 4708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebyf6FzKEiI.mp4", "duration": 115.2151, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the woman wearing a grey cardigan smiles while looking at the child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the woman wearing a grey cardigan smiles while looking at the child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, -0.37508195638656616], "gt_se_raw": [107.0, 114.0], "gt_se_norm": [0.8573954105377197, 0.9789072275161743], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004708.pt"}
{"idx": 4710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nG2rNBFzkGE.mp4", "duration": 16.616600000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the cat is looking out the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the cat is looking out the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.2778306007385254, 0.4443387985229492], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.5185537338256836, 0.3239772319793701], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004710.pt"}
{"idx": 4712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nG2rNBFzkGE.mp4", "duration": 16.616600000000002, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the cat is sitting still on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the cat is sitting still on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.3239772319793701], "gt_se_raw": [2.0, 10.0], "gt_se_norm": [-0.7592768669128418, 0.20361566543579102], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004712.pt"}
{"idx": 4714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uODXWDOj0ow.mp4", "duration": 30.797433333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the two men working with the yellow robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the two men working with the yellow robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5454167127609253], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5454167127609253], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004714.pt"}
{"idx": 4716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uODXWDOj0ow.mp4", "duration": 30.797433333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where three men are collaborating on a piece of machinery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where three men are collaborating on a piece of machinery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5454167127609253], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.09083336591720581, 0.36374998092651367], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004716.pt"}
{"idx": 4718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uODXWDOj0ow.mp4", "duration": 30.797433333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the blue and white striped shirt turns a red dial on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the blue and white striped shirt turns a red dial on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.220714271068573, -0.09083336591720581], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [-0.1557738184928894, -0.09083336591720581], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004718.pt"}
{"idx": 4720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/osP9iJjvlAE.mp4", "duration": 159.20070833333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a woman teaches another woman how to aim a pistol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a woman teaches another woman how to aim a pistol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.2211090326309204, -0.18342077732086182], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.20854628086090088, -0.18342077732086182], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004720.pt"}
{"idx": 4722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/osP9iJjvlAE.mp4", "duration": 159.20070833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the time when a group of people are running through the woods during the day.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when a group of people are running through the woods during the day.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 102.0], "base_se_norm": [0.23115026950836182, 0.28140127658843994], "gt_se_raw": [99.0, 101.0], "gt_se_norm": [0.24371302127838135, 0.2688385248184204], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004722.pt"}
{"idx": 4724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x-wcNG_Njk8.mp4", "duration": 126.29283333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the red clouds visible in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the red clouds visible in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 84.0], "base_se_norm": [-1.0, 0.3302417993545532], "gt_se_raw": [5.0, 81.0], "gt_se_norm": [-0.9208189249038696, 0.28273308277130127], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004724.pt"}
{"idx": 4726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x-wcNG_Njk8.mp4", "duration": 126.29283333333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a voice says, 'I never authorized any scarlet upgrade.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a voice says, 'I never authorized any scarlet upgrade.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.6040947437286377, -0.5249136686325073], "gt_se_raw": [81.0, 83.0], "gt_se_norm": [0.28273308277130127, 0.31440556049346924], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004726.pt"}
{"idx": 4728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x-wcNG_Njk8.mp4", "duration": 126.29283333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the time range that displays the social media information.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range that displays the social media information.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 126.0], "base_se_norm": [0.6944746971130371, 0.9953626394271851], "gt_se_raw": [107.0, 125.0], "gt_se_norm": [0.6944746971130371, 0.9795264005661011], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004728.pt"}
{"idx": 4730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RfI7lW6MbEU.mp4", "duration": 21.788433333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman in a red top holding a green cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman in a red top holding a green cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.2656654119491577, -0.08208179473876953], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.2656654119491577, -0.08208179473876953], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004730.pt"}
{"idx": 4732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RfI7lW6MbEU.mp4", "duration": 21.788433333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman scrape ingredients from a green cutting board into a pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman scrape ingredients from a green cutting board into a pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [0.2850855588912964, 0.46866917610168457], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [0.2850855588912964, 0.46866917610168457], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004732.pt"}
{"idx": 4734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZPkuz0gRPU.mp4", "duration": 156.86504166666666, "prompt_text": "Please find the visual event described by the sentence 'When are two women taking a selfie together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women taking a selfie together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [-0.5792561769485474, -0.5155071020126343], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.5920059680938721, -0.5155071020126343], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004734.pt"}
{"idx": 4736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZPkuz0gRPU.mp4", "duration": 156.86504166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a pink shirt uses a hand sanitizer dispenser.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a pink shirt uses a hand sanitizer dispenser.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 91.0], "base_se_norm": [0.10923385620117188, 0.1602330207824707], "gt_se_raw": [87.0, 91.0], "gt_se_norm": [0.10923385620117188, 0.1602330207824707], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004736.pt"}
{"idx": 4738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZPkuz0gRPU.mp4", "duration": 156.86504166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a black shirt take food from the pantry shelves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a black shirt take food from the pantry shelves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [-0.29876023530960083, -0.26051080226898193], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.29876023530960083, -0.26051080226898193], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004738.pt"}
{"idx": 4740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFinNxS5KN4.mp4", "duration": 160.11829166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a mosasaurus leaps out of the water to eat a shark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a mosasaurus leaps out of the water to eat a shark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 70.0], "base_se_norm": [-0.16311872005462646, -0.1256464123725891], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [-0.17560946941375732, -0.13813716173171997], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004740.pt"}
{"idx": 4742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFinNxS5KN4.mp4", "duration": 160.11829166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are people kayaking down a river surrounded by dinosaurs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people kayaking down a river surrounded by dinosaurs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 49.0], "base_se_norm": [-0.41293400526046753, -0.3879525065422058], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.41293400526046753, -0.3754616975784302], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004742.pt"}
{"idx": 4744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fDDw7W2RQYo.mp4", "duration": 150.98416666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point are two people communicating by touching hands on a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two people communicating by touching hands on a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.28469318151474, -0.2449539303779602], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.28469318151474, -0.2449539303779602], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004744.pt"}
{"idx": 4746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fDDw7W2RQYo.mp4", "duration": 150.98416666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the man and woman walking together on the sidewalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man and woman walking together on the sidewalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 92.0], "base_se_norm": [0.16568517684936523, 0.21867084503173828], "gt_se_raw": [90.0, 92.0], "gt_se_norm": [0.19217801094055176, 0.21867084503173828], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004746.pt"}
{"idx": 4748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fDDw7W2RQYo.mp4", "duration": 150.98416666666665, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when someone is swimming underwater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when someone is swimming underwater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 143.0], "base_se_norm": [0.5763242244720459, 0.8942384719848633], "gt_se_raw": [120.0, 134.0], "gt_se_norm": [0.5895707607269287, 0.7750205993652344], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004748.pt"}
{"idx": 4750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CzNb3zMq6C4.mp4", "duration": 185.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man in white sunglasses is dancing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man in white sunglasses is dancing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.6552962064743042, -0.6229802370071411], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.6552962064743042, -0.6229802370071411], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004750.pt"}
{"idx": 4752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CzNb3zMq6C4.mp4", "duration": 185.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man in the white sunglasses gives another member a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man in the white sunglasses gives another member a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [-0.37522441148757935, -0.32136446237564087], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.35368043184280396, -0.32136446237564087], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004752.pt"}
{"idx": 4754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsOzAbUt8n8.mp4", "duration": 124.68, "prompt_text": "Please find the visual event described by the sentence 'When is the title 'HIGHGROUND ADVENTURES' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the title 'HIGHGROUND ADVENTURES' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9679178595542908], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9679178595542908], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004754.pt"}
{"idx": 4756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsOzAbUt8n8.mp4", "duration": 124.68, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where a man in a pink shirt takes a selfie before the ride.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where a man in a pink shirt takes a selfie before the ride.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.18190568685531616, -0.1177414059638977], "gt_se_raw": [51.0, 55.0], "gt_se_norm": [-0.18190568685531616, -0.1177414059638977], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004756.pt"}
{"idx": 4758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsOzAbUt8n8.mp4", "duration": 124.68, "prompt_text": "Please find the visual event described by the sentence 'Locate the first-person perspective of someone flying over the river and green hills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the first-person perspective of someone flying over the river and green hills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 96.0], "base_se_norm": [0.15495669841766357, 0.5399422645568848], "gt_se_raw": [87.0, 91.0], "gt_se_norm": [0.3955726623535156, 0.4597369432449341], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004758.pt"}
{"idx": 4760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9fjEsxhmQi0.mp4", "duration": 79.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when a baby is sleeping in a white crib.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when a baby is sleeping in a white crib.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 79.0], "base_se_norm": [-1.0, 0.9758232831954956], "gt_se_raw": [1.0, 79.0], "gt_se_norm": [-0.9749895930290222, 0.9758232831954956], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004760.pt"}
{"idx": 4762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9fjEsxhmQi0.mp4", "duration": 79.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the interval where the background is pink with light effects.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the interval where the background is pink with light effects.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 79.0], "base_se_norm": [-0.9749895930290222, 0.9758232831954956], "gt_se_raw": [1.0, 79.0], "gt_se_norm": [-0.9749895930290222, 0.9758232831954956], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004762.pt"}
{"idx": 4764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5bWPhOp8TCw.mp4", "duration": 186.25273333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man get into the car and put on his seatbelt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get into the car and put on his seatbelt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 37.0], "base_se_norm": [-0.7530237436294556, -0.6026903986930847], "gt_se_raw": [23.0, 37.0], "gt_se_norm": [-0.7530237436294556, -0.6026903986930847], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004764.pt"}
{"idx": 4766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5bWPhOp8TCw.mp4", "duration": 186.25273333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man suggests the woman needs a makeover?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man suggests the woman needs a makeover?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [-0.30202364921569824, -0.22685694694519043], "gt_se_raw": [82.0, 85.0], "gt_se_norm": [-0.11947602033615112, -0.08726173639297485], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004766.pt"}
{"idx": 4768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5bWPhOp8TCw.mp4", "duration": 186.25273333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man finally get out of the car at his house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man finally get out of the car at his house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 157.0], "base_se_norm": [0.6536669731140137, 0.6858813762664795], "gt_se_raw": [151.0, 156.0], "gt_se_norm": [0.6214526891708374, 0.6751432418823242], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004768.pt"}
{"idx": 4770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d-84UJpYFRM.mp4", "duration": 198.16463333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the rat chases the researcher's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the rat chases the researcher's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 98.0], "base_se_norm": [-0.06138646602630615, -0.010923385620117188], "gt_se_raw": [92.0, 97.0], "gt_se_norm": [-0.07147908210754395, -0.02101600170135498], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004770.pt"}
{"idx": 4772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d-84UJpYFRM.mp4", "duration": 198.16463333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the researcher tickle the rat's belly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the researcher tickle the rat's belly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 143.0], "base_se_norm": [0.3927813768386841, 0.44324445724487305], "gt_se_raw": [139.0, 143.0], "gt_se_norm": [0.4028739929199219, 0.44324445724487305], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004772.pt"}
{"idx": 4774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gj6V-xZgtlQ.mp4", "duration": 319.7083333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the boy climbs up the shelves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the boy climbs up the shelves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 88.0], "base_se_norm": [-0.47452104091644287, -0.4494982361793518], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [-0.47452104091644287, -0.45575398206710815], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004774.pt"}
{"idx": 4776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gj6V-xZgtlQ.mp4", "duration": 319.7083333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the boy and his grandpa flying the red airplane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 192.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the boy and his grandpa flying the red airplane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 192.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 192.0], "base_se_norm": [0.10725915431976318, 0.20109474658966064], "gt_se_raw": [177.0, 187.0], "gt_se_norm": [0.10725915431976318, 0.169816255569458], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004776.pt"}
{"idx": 4778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gj6V-xZgtlQ.mp4", "duration": 319.7083333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where they are chased by aliens in space.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 222.0 - 247.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where they are chased by aliens in space.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 222.0 - 247.0 seconds.<|im_end|>\n", "base_se_raw": [222.0, 247.0], "base_se_norm": [0.388765811920166, 0.5451582670211792], "gt_se_raw": [221.0, 227.0], "gt_se_norm": [0.38251006603240967, 0.42004430294036865], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004778.pt"}
{"idx": 4780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q-FHWX1es0o.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the speaker compares the employment rates of people with disabilities to those without.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the speaker compares the employment rates of people with disabilities to those without.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7090288400650024, -0.4180576801300049], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004780.pt"}
{"idx": 4782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q-FHWX1es0o.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the bar chart for the year 1991 is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the bar chart for the year 1991 is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [78, 81], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004782.pt"}
{"idx": 4784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9lZlt-SlABw.mp4", "duration": 100.3002, "prompt_text": "Please find the visual event described by the sentence 'When is the woman initially presenting the beige vest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman initially presenting the beige vest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 16.0], "base_se_norm": [-0.8803591728210449, -0.6809577941894531], "gt_se_raw": [6.0, 17.0], "gt_se_norm": [-0.8803591728210449, -0.661017656326294], "start_token_span": [70, 73], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004784.pt"}
{"idx": 4786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9lZlt-SlABw.mp4", "duration": 100.3002, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman attach the red tinsel and run her fingers through it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman attach the red tinsel and run her fingers through it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 63.0], "base_se_norm": [-0.02293318510055542, 0.25622880458831787], "gt_se_raw": [48.0, 57.0], "gt_se_norm": [-0.0428733229637146, 0.1365879774093628], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004786.pt"}
{"idx": 4788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9lZlt-SlABw.mp4", "duration": 100.3002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the speaker explains that a child can categorize items as 'stringy' or 'shiny'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the speaker explains that a child can categorize items as 'stringy' or 'shiny'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 75.0], "base_se_norm": [0.25622880458831787, 0.495510458946228], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [0.35592949390411377, 0.4356900453567505], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004788.pt"}
{"idx": 4790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TdUsyXQ8Wrs.mp4", "duration": 115.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the movie title 'THE BLOB' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the movie title 'THE BLOB' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.6191575527191162, -0.5672245025634766], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.6191575527191162, -0.5672245025634766], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004790.pt"}
{"idx": 4792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TdUsyXQ8Wrs.mp4", "duration": 115.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the old man discover the blob on a stick and it attaches to his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the old man discover the blob on a stick and it attaches to his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 74.0], "base_se_norm": [0.15983843803405762, 0.28101563453674316], "gt_se_raw": [68.0, 74.0], "gt_se_norm": [0.17714941501617432, 0.28101563453674316], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004792.pt"}
{"idx": 4794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PquDhG2AhwM.mp4", "duration": 47.347300000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the child try to reach for the green shiny object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child try to reach for the green shiny object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, -0.2819020748138428], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.49310731887817383, -0.2819020748138428], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004794.pt"}
{"idx": 4796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PquDhG2AhwM.mp4", "duration": 47.347300000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child first touch the green tinsel on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child first touch the green tinsel on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.028455734252929688, 0.09826743602752686], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [-0.028455734252929688, 0.2249906063079834], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004796.pt"}
{"idx": 4798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PquDhG2AhwM.mp4", "duration": 47.347300000000004, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the child successfully grasps the tinsel with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the child successfully grasps the tinsel with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 42.0], "base_se_norm": [0.5206780433654785, 0.7741243839263916], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [0.7318832874298096, 0.9008475542068481], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004798.pt"}
{"idx": 4800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUjV78Kj9kQ.mp4", "duration": 87.52076666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment Snoopy is sharpening his knife and fork.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment Snoopy is sharpening his knife and fork.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 44.0], "base_se_norm": [-0.3373001217842102, 0.005475640296936035], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.3373001217842102, -0.24589323997497559], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004800.pt"}
{"idx": 4802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUjV78Kj9kQ.mp4", "duration": 87.52076666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when Snoopy and Woodstock pull the wishbone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when Snoopy and Woodstock pull the wishbone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 80.0], "base_se_norm": [0.7595824003219604, 0.8281375169754028], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [0.7595824003219604, 0.8052858114242554], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004802.pt"}
{"idx": 4804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-bCmuE3KxWI.mp4", "duration": 108.5, "prompt_text": "Please find the visual event described by the sentence 'When is the egg cracked into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the egg cracked into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [-0.7603686451911926, -0.7235022783279419], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.7603686451911926, -0.7235022783279419], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004804.pt"}
{"idx": 4806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-bCmuE3KxWI.mp4", "duration": 108.5, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dough kneaded by hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dough kneaded by hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [-0.023041486740112305, 0.06912446022033691], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.004608273506164551, 0.06912446022033691], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004806.pt"}
{"idx": 4808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-bCmuE3KxWI.mp4", "duration": 108.5, "prompt_text": "Please find the visual event described by the sentence 'When are the cookies placed into the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the cookies placed into the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [0.4930875301361084, 0.5483870506286621], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.4930875301361084, 0.5483870506286621], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004808.pt"}
{"idx": 4810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dTqW89AjVA0.mp4", "duration": 68.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the teacher is using sign language to explain the instructions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the teacher is using sign language to explain the instructions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 25.0], "base_se_norm": [-0.7665369510650635, -0.27042800188064575], "gt_se_raw": [8.0, 20.0], "gt_se_norm": [-0.7665369510650635, -0.41634243726730347], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004810.pt"}
{"idx": 4812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dTqW89AjVA0.mp4", "duration": 68.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the boy writes 'RS' inside the square.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the boy writes 'RS' inside the square.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.09533071517944336, -0.007782101631164551], "gt_se_raw": [44.0, 53.0], "gt_se_norm": [0.28404664993286133, 0.5466926097869873], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004812.pt"}
{"idx": 4814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MJesmjQXqno.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'When are the two medical professionals arranging surgical tools on a tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two medical professionals arranging surgical tools on a tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.6270396113395691], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.6803196668624878], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004814.pt"}
{"idx": 4816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MJesmjQXqno.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'At what point is the median salary of $38,838 displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the median salary of $38,838 displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.20079916715621948, 0.11888110637664795], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.20079916715621948, 0.11888110637664795], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004816.pt"}
{"idx": 4818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MJesmjQXqno.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the text 'YOU'VE GOT THIS!' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the text 'YOU'VE GOT THIS!' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [0.5451215505599976, 0.8648018836975098], "gt_se_raw": [29.0, 35.0], "gt_se_norm": [0.5451215505599976, 0.8648018836975098], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004818.pt"}
{"idx": 4820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5WeiYHnvNY.mp4", "duration": 161.7, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a person is reading braille.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a person is reading braille.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 62.0], "base_se_norm": [-0.2702535390853882, -0.23314779996871948], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.2702535390853882, -0.23314779996871948], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004820.pt"}
{"idx": 4822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5WeiYHnvNY.mp4", "duration": 161.7, "prompt_text": "Please find the visual event described by the sentence 'Find the man in a suit and red tie as he delivers a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the man in a suit and red tie as he delivers a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 59.0], "base_se_norm": [-0.49288803339004517, -0.2702535390853882], "gt_se_raw": [41.0, 59.0], "gt_se_norm": [-0.49288803339004517, -0.2702535390853882], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004822.pt"}
{"idx": 4824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdDX_uhxSwg.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman with blonde hair picking up a glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman with blonde hair picking up a glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9257425665855408, -0.40594059228897095], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9257425665855408, -0.7772276997566223], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004824.pt"}
{"idx": 4826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdDX_uhxSwg.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the person in the tie-dye shirt puts their hand on the red object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the person in the tie-dye shirt puts their hand on the red object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 22.0], "base_se_norm": [-0.034653425216674805, 0.6336634159088135], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [0.03960394859313965, 0.2623763084411621], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004826.pt"}
{"idx": 4828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdDX_uhxSwg.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman on the left place both of her hands on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman on the left place both of her hands on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 22.0], "base_se_norm": [-0.034653425216674805, 0.6336634159088135], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [0.11386144161224365, 0.41089117527008057], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004828.pt"}
{"idx": 4830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VJwEh7xne-g.mp4", "duration": 64.24011951650142, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man in the pink shirt spreads his fingers over the woman's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man in the pink shirt spreads his fingers over the woman's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 45.0], "base_se_norm": [-0.06600421667098999, 0.4009937047958374], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.06600421667098999, 0.02739536762237549], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004830.pt"}
{"idx": 4832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VJwEh7xne-g.mp4", "duration": 64.24011951650142, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man raises both his hands in a gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man raises both his hands in a gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.4943932294845581, 0.6189260482788086], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.46326005458831787, 0.5255264043807983], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004832.pt"}
{"idx": 4834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtBBL7Pb9Q0.mp4", "duration": 43.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the yellow train approaching through the market?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the yellow train approaching through the market?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 30.0], "base_se_norm": [-0.8159509301185608, 0.38036811351776123], "gt_se_raw": [3.0, 16.0], "gt_se_norm": [-0.8619632124900818, -0.2638036608695984], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004834.pt"}
{"idx": 4836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtBBL7Pb9Q0.mp4", "duration": 43.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point do the vendors start putting their awnings and goods back over the tracks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the vendors start putting their awnings and goods back over the tracks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 42.0], "base_se_norm": [0.4723926782608032, 0.9325152635574341], "gt_se_raw": [30.0, 42.0], "gt_se_norm": [0.38036811351776123, 0.9325152635574341], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004836.pt"}
{"idx": 4838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtBBL7Pb9Q0.mp4", "duration": 43.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the train blow its horn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the train blow its horn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.3098159432411194, -0.17177915573120117], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.6779141426086426, -0.4018405079841614], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004838.pt"}
{"idx": 4840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lgbdir5RAHc.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man stands with his arms crossed while talking to the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man stands with his arms crossed while talking to the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 51.0], "base_se_norm": [-0.6470297574996948, -0.47054457664489746], "gt_se_raw": [34.0, 50.0], "gt_se_norm": [-0.6470297574996948, -0.4809260964393616], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004840.pt"}
{"idx": 4842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lgbdir5RAHc.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman slap the man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman slap the man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 83.0], "base_se_norm": [-0.1591002345085144, -0.13833731412887573], "gt_se_raw": [81.0, 82.0], "gt_se_norm": [-0.1591002345085144, -0.14871877431869507], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004842.pt"}
{"idx": 4844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6NCN13K7Js.mp4", "duration": 81.01426666666667, "prompt_text": "Please find the visual event described by the sentence 'When is a woman interacting with a child in a special chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman interacting with a child in a special chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8765649795532227, -0.7531298995018005], "gt_se_raw": [4.0, 10.0], "gt_se_norm": [-0.9012519717216492, -0.7531298995018005], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004844.pt"}
{"idx": 4846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6NCN13K7Js.mp4", "duration": 81.01426666666667, "prompt_text": "Please find the visual event described by the sentence 'When is a woman wearing a polka-dot scarf speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman wearing a polka-dot scarf speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 23.0], "base_se_norm": [-0.6050078868865967, -0.43219876289367676], "gt_se_raw": [16.0, 24.0], "gt_se_norm": [-0.6050078868865967, -0.40751177072525024], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004846.pt"}
{"idx": 4848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6NCN13K7Js.mp4", "duration": 81.01426666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a group of people are standing in front of a green wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a group of people are standing in front of a green wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 77.0], "base_se_norm": [0.6046556234359741, 0.9008997678756714], "gt_se_raw": [65.0, 77.0], "gt_se_norm": [0.6046556234359741, 0.9008997678756714], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004848.pt"}
{"idx": 4850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4mssLzXoSg.mp4", "duration": 141.76662499999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the driver is smoking a cigarette.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the driver is smoking a cigarette.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.8307076692581177, -0.717846155166626], "gt_se_raw": [18.0, 27.0], "gt_se_norm": [-0.7460615634918213, -0.6190923452377319], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004850.pt"}
{"idx": 4852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4mssLzXoSg.mp4", "duration": 141.76662499999998, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where the man in the trench coat is shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where the man in the trench coat is shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 100.0], "base_se_norm": [0.3120154142379761, 0.4107692241668701], "gt_se_raw": [88.0, 90.0], "gt_se_norm": [0.2414768934249878, 0.2696923017501831], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004852.pt"}
{"idx": 4854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0brZ8Cinao.mp4", "duration": 29.766666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a group of friends are sitting on the curb, laughing and enjoying their Cokes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a group of friends are sitting on the curb, laughing and enjoying their Cokes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.7984322309494019, -0.5968644618988037], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.8656215071678162, -0.5968644618988037], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004854.pt"}
{"idx": 4856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0brZ8Cinao.mp4", "duration": 29.766666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the green shirt enter the computer lab and turn on the lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the green shirt enter the computer lab and turn on the lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [-0.12653976678848267, 0.007838726043701172], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.12653976678848267, 0.007838726043701172], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004856.pt"}
{"idx": 4858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0brZ8Cinao.mp4", "duration": 29.766666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a bottle of Coca-Cola is opened and fizzes over, with the sun in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a bottle of Coca-Cola is opened and fizzes over, with the sun in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.5453526973724365, 0.7469204664230347], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [0.5453526973724365, 0.7469204664230347], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004858.pt"}
{"idx": 4860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34aAAxaOABU.mp4", "duration": 69.43603333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the person stand up and start building the torso?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person stand up and start building the torso?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 26.0], "base_se_norm": [-0.6255546808242798, -0.2511093020439148], "gt_se_raw": [16.0, 30.0], "gt_se_norm": [-0.5391441583633423, -0.13589537143707275], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004860.pt"}
{"idx": 4862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34aAAxaOABU.mp4", "duration": 69.43603333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the head is being added to the sculpture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the head is being added to the sculpture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 64.0], "base_se_norm": [-0.04948490858078003, 0.8434232473373413], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [0.3537639379501343, 0.4977813959121704], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004862.pt"}
{"idx": 4864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SZAraIESRCY.mp4", "duration": 143.91043333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the young boy in the red shirt interacting with the cardboard sculpture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the young boy in the red shirt interacting with the cardboard sculpture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.8888197541236877, -0.7776394486427307], "gt_se_raw": [8.0, 17.0], "gt_se_norm": [-0.8888197541236877, -0.7637419700622559], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004864.pt"}
{"idx": 4866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SZAraIESRCY.mp4", "duration": 143.91043333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a barber cutting a man's hair in a barbershop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a barber cutting a man's hair in a barbershop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 65.0], "base_se_norm": [-0.26343071460723877, -0.096660315990448], "gt_se_raw": [53.0, 65.0], "gt_se_norm": [-0.26343071460723877, -0.096660315990448], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004866.pt"}
{"idx": 4868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SZAraIESRCY.mp4", "duration": 143.91043333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the older woman with curly hair and a striped shirt speaking into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the older woman with curly hair and a striped shirt speaking into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 115.0], "base_se_norm": [0.37585580348968506, 0.5982162952423096], "gt_se_raw": [100.0, 115.0], "gt_se_norm": [0.3897533416748047, 0.5982162952423096], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004868.pt"}
{"idx": 4870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ADrkVCYoR2U.mp4", "duration": 43.00963333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman rocking the baby in the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman rocking the baby in the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.5349879264831543], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [0.20903146266937256, 0.39503633975982666], "start_token_span": [71, 74], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004870.pt"}
{"idx": 4872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ADrkVCYoR2U.mp4", "duration": 43.00963333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do two adults appear and start a conversation against a blue background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two adults appear and start a conversation against a blue background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [0.39503633975982666, 0.5345399379730225], "gt_se_raw": [31.0, 32.0], "gt_se_norm": [0.4415374994277954, 0.4880387783050537], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004872.pt"}
{"idx": 4874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/peu8CIu1W8Q.mp4", "duration": 91.45803333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman in a purple shirt speaking? please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman in a purple shirt speaking? please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.8906602263450623, -0.7375845909118652], "gt_se_raw": [4.0, 12.0], "gt_se_norm": [-0.9125281572341919, -0.7375845909118652], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004874.pt"}
{"idx": 4876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/peu8CIu1W8Q.mp4", "duration": 91.45803333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a woman with short brown hair introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a woman with short brown hair introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.43143320083618164, -0.30022549629211426], "gt_se_raw": [26.0, 32.0], "gt_se_norm": [-0.43143320083618164, -0.30022549629211426], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004876.pt"}
{"idx": 4878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/peu8CIu1W8Q.mp4", "duration": 91.45803333333333, "prompt_text": "Please find the visual event described by the sentence 'When are people clapping their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people clapping their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [0.2027374505996704, 0.2683413028717041], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [0.22460544109344482, 0.2683413028717041], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004878.pt"}
{"idx": 4880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgZBmmuB_wg.mp4", "duration": 19.583333333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the character with a TV head appears and approaches the judge's bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the character with a TV head appears and approaches the judge's bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.5914893746376038, -0.2851064205169678], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.5914893746376038, -0.387234091758728], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004880.pt"}
{"idx": 4882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgZBmmuB_wg.mp4", "duration": 19.583333333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me when the judge picks up the gavel and strikes it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the judge picks up the gavel and strikes it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [0.3276594877243042, 0.634042501449585], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [0.22553181648254395, 0.5319148302078247], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004882.pt"}
{"idx": 4884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1PhPYr_9zRY.mp4", "duration": 233.35812499999997, "prompt_text": "Please find the visual event described by the sentence 'When can we see the men clapping their hands with excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the men clapping their hands with excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9914295077323914, -0.9742884635925293], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9828589558601379], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004884.pt"}
{"idx": 4886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1PhPYr_9zRY.mp4", "duration": 233.35812499999997, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the blue shirt look surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the blue shirt look surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7857370376586914, -0.7600255012512207], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9400063753128052, -0.9228653311729431], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004886.pt"}
{"idx": 4888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1PhPYr_9zRY.mp4", "duration": 233.35812499999997, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man in a light yellow shirt raises his right hand to high-five another man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 206.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man in a light yellow shirt raises his right hand to high-five another man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 206.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 206.0], "base_se_norm": [0.7483856678009033, 0.7655266523361206], "gt_se_raw": [150.0, 152.0], "gt_se_norm": [0.285577654838562, 0.30271875858306885], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004888.pt"}
{"idx": 4890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RqWYIB1O-0.mp4", "duration": 27.8, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman uses the iPad to communicate with her colleague.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman uses the iPad to communicate with her colleague.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 20.0], "base_se_norm": [-0.2805755138397217, 0.43884897232055664], "gt_se_raw": [11.0, 21.0], "gt_se_norm": [-0.20863306522369385, 0.5107914209365845], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004890.pt"}
{"idx": 4892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RqWYIB1O-0.mp4", "duration": 27.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the woman in the red shirt turns and walks out the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the woman in the red shirt turns and walks out the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [0.726618766784668, 0.9424461126327515], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.726618766784668, 0.8705036640167236], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004892.pt"}
{"idx": 4894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8fn4fewrvQg.mp4", "duration": 72.53912245578913, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person in a red jacket opens a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person in a red jacket opens a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 26.0], "base_se_norm": [-0.4210020899772644, -0.2831454873085022], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.4210020899772644, -0.2831454873085022], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004894.pt"}
{"idx": 4896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8fn4fewrvQg.mp4", "duration": 72.53912245578913, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man putting a striped bag into the trunk of a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man putting a striped bag into the trunk of a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 39.0], "base_se_norm": [-0.06257480382919312, 0.07528185844421387], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [-0.09014618396759033, 0.020139098167419434], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004896.pt"}
{"idx": 4898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8fn4fewrvQg.mp4", "duration": 72.53912245578913, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the 'Heritage Hall' sign is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the 'Heritage Hall' sign is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 65.0], "base_se_norm": [0.736993670463562, 0.7921363115310669], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [0.7094223499298096, 0.7921363115310669], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004898.pt"}
{"idx": 4900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cB7O_JIxGuY.mp4", "duration": 126.4263, "prompt_text": "Please find the visual event described by the sentence 'At what point does the poison apple take on the appearance of a skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the poison apple take on the appearance of a skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.7468881011009216, -0.6836101412773132], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.7468881011009216, -0.6836101412773132], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004900.pt"}
{"idx": 4902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cB7O_JIxGuY.mp4", "duration": 126.4263, "prompt_text": "Please find the visual event described by the sentence 'When can we see the old witch peeking through the cottage window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the old witch peeking through the cottage window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 75.0], "base_se_norm": [0.0282670259475708, 0.18646204471588135], "gt_se_raw": [66.0, 72.0], "gt_se_norm": [0.044086575508117676, 0.13900351524353027], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004902.pt"}
{"idx": 4904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0JboM-STb4E.mp4", "duration": 151.23946248226358, "prompt_text": "Please find the visual event described by the sentence 'When do two women toast with red cups?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two women toast with red cups?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.9338796734809875, -0.8809834718704224], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.94710373878479, -0.9206556081771851], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004904.pt"}
{"idx": 4906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0JboM-STb4E.mp4", "duration": 151.23946248226358, "prompt_text": "Please find the visual event described by the sentence 'When do two puppies kiss each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two puppies kiss each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 46.0], "base_se_norm": [-0.41814124584198, -0.391693115234375], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.4049171805381775, -0.3784690499305725], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004906.pt"}
{"idx": 4908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0JboM-STb4E.mp4", "duration": 151.23946248226358, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman is surrounded by a large group of puppies on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman is surrounded by a large group of puppies on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 97.0], "base_se_norm": [0.16371750831604004, 0.2827340364456177], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.24306190013885498, 0.2827340364456177], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004908.pt"}
{"idx": 4910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24Pmmo9wKik.mp4", "duration": 70.87100000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a black leather jacket walks into a tent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a black leather jacket walks into a tent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.5766956806182861, -0.43559426069259644], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.46381455659866333, -0.40737396478652954], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004910.pt"}
{"idx": 4912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24Pmmo9wKik.mp4", "duration": 70.87100000000001, "prompt_text": "Please find the visual event described by the sentence 'When is the campfire put out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the campfire put out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 48.0], "base_se_norm": [0.01593029499053955, 0.35457372665405273], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [0.2699129581451416, 0.32635343074798584], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004912.pt"}
{"idx": 4914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNi_HC839Wo.mp4", "duration": 153.06957777777777, "prompt_text": "Please find the visual event described by the sentence 'When do the girl and her mother have a major argument in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the girl and her mother have a major argument in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.8432085514068604, -0.7125490307807922], "gt_se_raw": [2.0, 22.0], "gt_se_norm": [-0.9738680720329285, -0.7125490307807922], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004914.pt"}
{"idx": 4916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNi_HC839Wo.mp4", "duration": 153.06957777777777, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the mother and daughter shop for a prom dress at a thrift store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the mother and daughter shop for a prom dress at a thrift store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 73.0], "base_se_norm": [-0.1507130265235901, -0.04618537425994873], "gt_se_raw": [65.0, 73.0], "gt_se_norm": [-0.1507130265235901, -0.04618537425994873], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004916.pt"}
{"idx": 4918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNi_HC839Wo.mp4", "duration": 153.06957777777777, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the girl arguing with her friends in the schoolyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the girl arguing with her friends in the schoolyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 117.0], "base_se_norm": [0.3849910497665405, 0.5287165641784668], "gt_se_raw": [107.0, 117.0], "gt_se_norm": [0.3980569839477539, 0.5287165641784668], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004918.pt"}
{"idx": 4920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hkmvuV6PK20.mp4", "duration": 162.82933333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy rub his eyes after crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy rub his eyes after crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [-0.4718396067619324, -0.4349912405014038], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.4718396067619324, -0.4227084517478943], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004920.pt"}
{"idx": 4922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hkmvuV6PK20.mp4", "duration": 162.82933333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a horse-drawn cart is visible outside the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a horse-drawn cart is visible outside the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 101.0], "base_se_norm": [0.17914879322052002, 0.2405627965927124], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [0.19143152236938477, 0.2282799482345581], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004922.pt"}
{"idx": 4924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZEnkdcZuhnQ.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9469613432884216], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.960220992565155, -0.9337016344070435], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004924.pt"}
{"idx": 4926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZEnkdcZuhnQ.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the speaker discusses providing resources such as housing and transportation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the speaker discusses providing resources such as housing and transportation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 50.0], "base_se_norm": [-0.5226519107818604, -0.3370165228843689], "gt_se_raw": [34.0, 47.0], "gt_se_norm": [-0.5491712689399719, -0.37679553031921387], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004926.pt"}
{"idx": 4928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZEnkdcZuhnQ.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the part where she explains how often consumers are eligible for a new device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where she explains how often consumers are eligible for a new device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [0.3259669542312622, 0.48508286476135254], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.1911602020263672, -0.13812154531478882], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004928.pt"}
{"idx": 4930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fj3u4Wd2C-8.mp4", "duration": 155.48866666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker explain the 'Dwell' function for clicking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker explain the 'Dwell' function for clicking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 108.0], "base_se_norm": [0.3505808115005493, 0.3891688585281372], "gt_se_raw": [106.0, 109.0], "gt_se_norm": [0.3634434938430786, 0.40203142166137695], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004930.pt"}
{"idx": 4932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fj3u4Wd2C-8.mp4", "duration": 155.48866666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the video explain how the eye-tracker's cameras work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video explain how the eye-tracker's cameras work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 90.0], "base_se_norm": [-0.07388746738433838, 0.15764069557189941], "gt_se_raw": [80.0, 83.0], "gt_se_norm": [0.029013872146606445, 0.06760191917419434], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004932.pt"}
{"idx": 4934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F-eMt3SrfFU.mp4", "duration": 138.7636222222222, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a soldier walks towards and into the sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a soldier walks towards and into the sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.7549790143966675, -0.6973270177841187], "gt_se_raw": [44.0, 50.0], "gt_se_norm": [-0.3658280372619629, -0.27935004234313965], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004934.pt"}
{"idx": 4936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F-eMt3SrfFU.mp4", "duration": 138.7636222222222, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a large group of soldiers is standing in line on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a large group of soldiers is standing in line on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6108490228652954, -0.5676100254058838], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.6108490228652954, -0.582023024559021], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004936.pt"}
{"idx": 4938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F-eMt3SrfFU.mp4", "duration": 138.7636222222222, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a bomb explodes on the beach, sending sand flying everywhere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a bomb explodes on the beach, sending sand flying everywhere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.48113203048706055, -0.4234800338745117], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.48113203048706055, -0.4378930330276489], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004938.pt"}
{"idx": 4940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCdx7-XOmVc.mp4", "duration": 222.48893333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker list various slang terms for money, from 'wealth' to 'greenbacks'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker list various slang terms for money, from 'wealth' to 'greenbacks'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 74.0], "base_se_norm": [-0.5325610637664795, -0.3347983956336975], "gt_se_raw": [5.0, 14.0], "gt_se_norm": [-0.9550539255142212, -0.8741510510444641], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004940.pt"}
{"idx": 4942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCdx7-XOmVc.mp4", "duration": 222.48893333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the speaker holds up a fan of dollar bills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 212.0 - 215.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the speaker holds up a fan of dollar bills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 212.0 - 215.0 seconds.<|im_end|>\n", "base_se_raw": [212.0, 215.0], "base_se_norm": [0.9057127237319946, 0.9326803684234619], "gt_se_raw": [212.0, 215.0], "gt_se_norm": [0.9057127237319946, 0.9326803684234619], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004942.pt"}
{"idx": 4944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NVGGgncVq-4.mp4", "duration": 253.24, "prompt_text": "Please find the visual event described by the sentence 'When are the children playing in the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children playing in the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, -0.8499447107315063], "gt_se_raw": [0.0, 19.0], "gt_se_norm": [-1.0, -0.8499447107315063], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004944.pt"}
{"idx": 4946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NVGGgncVq-4.mp4", "duration": 253.24, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the father tucks his daughter into bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the father tucks his daughter into bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 55.0], "base_se_norm": [-0.605117678642273, -0.5656294822692871], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.5972200632095337, -0.5577318072319031], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004946.pt"}
{"idx": 4948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NVGGgncVq-4.mp4", "duration": 253.24, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a man and a girl are browsing a website on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 209.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a man and a girl are browsing a website on a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 209.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 209.0], "base_se_norm": [0.3347022533416748, 0.6506080627441406], "gt_se_raw": [170.0, 209.0], "gt_se_norm": [0.34259986877441406, 0.6506080627441406], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004948.pt"}
{"idx": 4950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mgNgscHJh6I.mp4", "duration": 67.06700000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the white coat walks away from the coffin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the white coat walks away from the coffin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 24.0], "base_se_norm": [-0.49304425716400146, -0.2842978239059448], "gt_se_raw": [18.0, 23.0], "gt_se_norm": [-0.46322333812713623, -0.31411874294281006], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004950.pt"}
{"idx": 4952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mgNgscHJh6I.mp4", "duration": 67.06700000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman opens the envelope and starts to look at the letter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman opens the envelope and starts to look at the letter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 44.0], "base_se_norm": [-0.045730412006378174, 0.3121206760406494], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [0.25247883796691895, 0.34194159507751465], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004952.pt"}
{"idx": 4954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdEfx8up7wM.mp4", "duration": 68.44337499999999, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the boy in the green jacket wearing a gas mask?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the boy in the green jacket wearing a gas mask?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.44479650259017944, -0.35713279247283936], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.5324602127075195, -0.4155752658843994], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004954.pt"}
{"idx": 4956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdEfx8up7wM.mp4", "duration": 68.44337499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a top-down view of the group sitting in a circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a top-down view of the group sitting in a circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.35713279247283936, -0.26946908235549927], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [-0.3863540291786194, -0.26946908235549927], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004956.pt"}
{"idx": 4958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdEfx8up7wM.mp4", "duration": 68.44337499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the can finally explode?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the can finally explode?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 64.0], "base_se_norm": [0.8117166757583618, 0.8701591491699219], "gt_se_raw": [64.0, 65.0], "gt_se_norm": [0.8701591491699219, 0.8993803262710571], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004958.pt"}
{"idx": 4960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvHSlHhh1gk.mp4", "duration": 144.14399999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a pink heart-shaped object appears and says 'Hello'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a pink heart-shaped object appears and says 'Hello'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 70.0], "base_se_norm": [-0.05649906396865845, -0.02874898910522461], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.05649906396865845, -0.02874898910522461], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004960.pt"}
{"idx": 4962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvHSlHhh1gk.mp4", "duration": 144.14399999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does the white robot with wings capture the group of heroes in a force field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the white robot with wings capture the group of heroes in a force field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.5956265926361084, 0.637251615524292], "gt_se_raw": [115.0, 117.0], "gt_se_norm": [0.5956265926361084, 0.6233766078948975], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004962.pt"}
{"idx": 4964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfQHp_px7iI.mp4", "duration": 32.16, "prompt_text": "Please find the visual event described by the sentence 'When are the text overlays describing the vacuum's features shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the text overlays describing the vacuum's features shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 8.0], "base_se_norm": [-0.8756219148635864, -0.5024875402450562], "gt_se_raw": [2.0, 8.0], "gt_se_norm": [-0.8756219148635864, -0.5024875402450562], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004964.pt"}
{"idx": 4966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfQHp_px7iI.mp4", "duration": 32.16, "prompt_text": "Please find the visual event described by the sentence 'At what point does a person press the red power button?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a person press the red power button?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.25373131036758423, -0.12935322523117065], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.25373131036758423, -0.12935322523117065], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004966.pt"}
{"idx": 4968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfQHp_px7iI.mp4", "duration": 32.16, "prompt_text": "Please find the visual event described by the sentence 'When is there a close-up shot of the 'dyson kinetic big ball animal' text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is there a close-up shot of the 'dyson kinetic big ball animal' text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.43034827709198, 0.7412935495376587], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.43034827709198, 0.7412935495376587], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004968.pt"}
{"idx": 4970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FuraQCCsKgE.mp4", "duration": 8.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the dog is playing with the orange ball and making it squeak.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the dog is playing with the orange ball and making it squeak.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.5256917476654053, 0.18577075004577637], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.5256917476654053, -0.28853756189346313], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004970.pt"}
{"idx": 4972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FuraQCCsKgE.mp4", "duration": 8.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the dog is howling loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the dog is howling loudly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [0.18577075004577637, 0.8972331285476685], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [0.18577075004577637, 0.6600790023803711], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004972.pt"}
{"idx": 4974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YEG08_h6aT8.mp4", "duration": 150.6, "prompt_text": "Please find the visual event described by the sentence 'When can we see two chipmunks on the branch of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two chipmunks on the branch of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 35.0], "base_se_norm": [-0.5883134603500366, -0.535192608833313], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.5883134603500366, -0.535192608833313], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004974.pt"}
{"idx": 4976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YEG08_h6aT8.mp4", "duration": 150.6, "prompt_text": "Please find the visual event described by the sentence 'When does the owl wake up at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the owl wake up at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 83.0], "base_se_norm": [-0.01726430654525757, 0.1022576093673706], "gt_se_raw": [73.0, 84.0], "gt_se_norm": [-0.030544519424438477, 0.11553776264190674], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004976.pt"}
{"idx": 4978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YEG08_h6aT8.mp4", "duration": 150.6, "prompt_text": "Please find the visual event described by the sentence 'At what time do two brown rabbits appear on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do two brown rabbits appear on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 116.0], "base_se_norm": [0.4475431442260742, 0.5405045747756958], "gt_se_raw": [110.0, 118.0], "gt_se_norm": [0.46082329750061035, 0.5670650005340576], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004978.pt"}
{"idx": 4980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2y1QQWNZxZM.mp4", "duration": 33.06636666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the father and the child are crawling together on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the father and the child are crawling together on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 23.0], "base_se_norm": [-0.8185467720031738, 0.39114153385162354], "gt_se_raw": [10.0, 24.0], "gt_se_norm": [-0.3951558470726013, 0.4516260623931885], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004980.pt"}
{"idx": 4982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2y1QQWNZxZM.mp4", "duration": 33.06636666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the child buries their face in the blue sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the child buries their face in the blue sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [0.69356369972229, 0.9355013370513916], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.7540481090545654, 0.9355013370513916], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004982.pt"}
{"idx": 4984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uy7ykWjFH1E.mp4", "duration": 143.6435, "prompt_text": "Please find the visual event described by the sentence 'When can we see an animated man running away from a fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see an animated man running away from a fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 23.0], "base_se_norm": [-0.7911496162414551, -0.6797627210617065], "gt_se_raw": [16.0, 23.0], "gt_se_norm": [-0.7772262692451477, -0.6797627210617065], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004984.pt"}
{"idx": 4986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uy7ykWjFH1E.mp4", "duration": 143.6435, "prompt_text": "Please find the visual event described by the sentence 'At what time is a photo of a classroom filled with children shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a photo of a classroom filled with children shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 81.0], "base_se_norm": [0.0024819374084472656, 0.1277921199798584], "gt_se_raw": [72.0, 81.0], "gt_se_norm": [0.0024819374084472656, 0.1277921199798584], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004986.pt"}
{"idx": 4988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uy7ykWjFH1E.mp4", "duration": 143.6435, "prompt_text": "Please find the visual event described by the sentence 'Find the image of colorful paper cut-out children holding hands, including some in wheelchairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the image of colorful paper cut-out children holding hands, including some in wheelchairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 143.0], "base_se_norm": [0.7682666778564453, 0.9910403490066528], "gt_se_raw": [127.0, 143.0], "gt_se_norm": [0.7682666778564453, 0.9910403490066528], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004988.pt"}
{"idx": 4990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wji-BZ0oCwg.mp4", "duration": 117.32554166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the giant spherical space station appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the giant spherical space station appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.16471725702285767, -0.13062405586242676], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.16471725702285767, -0.13062405586242676], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004990.pt"}
{"idx": 4992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wji-BZ0oCwg.mp4", "duration": 117.32554166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see a silhouette of a figure in a black cape appearing from a steamy doorway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see a silhouette of a figure in a black cape appearing from a steamy doorway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 79.0], "base_se_norm": [0.3125871419906616, 0.3466804027557373], "gt_se_raw": [78.0, 79.0], "gt_se_norm": [0.32963383197784424, 0.3466804027557373], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004992.pt"}
{"idx": 4994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KJiCalhfN6k.mp4", "duration": 22.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the green shirt touch the cards under the 'Monday' column?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the green shirt touch the cards under the 'Monday' column?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.11373710632324219, 0.24076807498931885], "gt_se_raw": [10.0, 18.0], "gt_se_norm": [-0.11373710632324219, 0.5952732563018799], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004994.pt"}
{"idx": 4996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KJiCalhfN6k.mp4", "duration": 22.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'home' card for Sunday being touched?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'home' card for Sunday being touched?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.3796159625053406, 0.24076807498931885], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.3796159625053406, -0.20236337184906006], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004996.pt"}
{"idx": 4998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KJiCalhfN6k.mp4", "duration": 22.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the 'vacation' card being pointed at?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'vacation' card being pointed at?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [0.24076807498931885, 0.8611520528793335], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [0.3293943405151367, 0.5952732563018799], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004998.pt"}
{"idx": 5000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/znxRGH92XDg.mp4", "duration": 157.6575, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman is running to catch the bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman is running to catch the bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 38.0], "base_se_norm": [-0.6574853658676147, -0.5179423689842224], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.6574853658676147, -0.5433138608932495], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005000.pt"}
{"idx": 5002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/znxRGH92XDg.mp4", "duration": 157.6575, "prompt_text": "Please find the visual event described by the sentence 'When does the man in overalls show the woman the gold powder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in overalls show the woman the gold powder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 73.0], "base_se_norm": [-0.18811345100402832, -0.07394194602966309], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [-0.162742018699646, -0.12468487024307251], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005002.pt"}
{"idx": 5004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ate2ecHwBoo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman leading a black horse in an indoor arena?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman leading a black horse in an indoor arena?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005004.pt"}
{"idx": 5006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ate2ecHwBoo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a black jacket with 'MALIBU MOON' written on it presenting a horse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a black jacket with 'MALIBU MOON' written on it presenting a horse?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.08188092708587646, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005006.pt"}
{"idx": 5008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ate2ecHwBoo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the text 'EQUINE STUDIES' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the text 'EQUINE STUDIES' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9344200491905212, -0.5409404635429382], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9344200491905212, -0.5409404635429382], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005008.pt"}
{"idx": 5010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2qaNcyIoLV8.mp4", "duration": 132.76596666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip of the man in the wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip of the man in the wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 86.0], "base_se_norm": [0.09967947006225586, 0.2955127954483032], "gt_se_raw": [73.0, 87.0], "gt_se_norm": [0.09967947006225586, 0.3105769157409668], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005010.pt"}
{"idx": 5012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2qaNcyIoLV8.mp4", "duration": 132.76596666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman in a red jacket and sunglasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman in a red jacket and sunglasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 73.0], "base_se_norm": [-0.15641027688980103, 0.09967947006225586], "gt_se_raw": [57.0, 73.0], "gt_se_norm": [-0.14134615659713745, 0.09967947006225586], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005012.pt"}
{"idx": 5014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R_j_5joOBL0.mp4", "duration": 126.7266, "prompt_text": "Please find the visual event described by the sentence 'When is a cup of juice placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a cup of juice placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 22.0], "base_se_norm": [-0.6843599081039429, -0.6527958512306213], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.7001419067382812, -0.6527958512306213], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005014.pt"}
{"idx": 5016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R_j_5joOBL0.mp4", "duration": 126.7266, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man wearing a red jersey is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man wearing a red jersey is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.905307948589325, -0.8421799540519714], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.4791938066482544, -0.4002837538719177], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005016.pt"}
{"idx": 5018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R_j_5joOBL0.mp4", "duration": 126.7266, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man try to open the bottle with his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man try to open the bottle with his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.5581037998199463, -0.4949758052825928], "gt_se_raw": [124.0, 126.0], "gt_se_norm": [0.9569687843322754, 0.9885327816009521], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005018.pt"}
{"idx": 5020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aI51UWF8_9Y.mp4", "duration": 115.98253333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where three people are dancing in a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where three people are dancing in a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.24126505851745605, -0.1722891926765442], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.24126505851745605, -0.1722891926765442], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005020.pt"}
{"idx": 5022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aI51UWF8_9Y.mp4", "duration": 115.98253333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman is doing a headstand on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman is doing a headstand on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 91.0], "base_se_norm": [0.5002259016036987, 0.5692017078399658], "gt_se_raw": [87.0, 90.0], "gt_se_norm": [0.5002259016036987, 0.5519577264785767], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005022.pt"}
{"idx": 5024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6pBONClIuzg.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man walking on a street at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man walking on a street at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.5228828191757202, -0.45219874382019043], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.5228828191757202, -0.45219874382019043], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005024.pt"}
{"idx": 5026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6pBONClIuzg.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the Marvel logo displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the Marvel logo displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.6642508506774902, -0.6112378239631653], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.6465798616409302, -0.6112378239631653], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005026.pt"}
{"idx": 5028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6pBONClIuzg.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with blonde hair walk down a hospital corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with blonde hair walk down a hospital corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [-0.11644959449768066, -0.06343656778335571], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.11644959449768066, -0.08110755681991577], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005028.pt"}
{"idx": 5030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ck2K69zZBG4.mp4", "duration": 136.42795833333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man is pointing at his tie while speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man is pointing at his tie while speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 52.0], "base_se_norm": [-0.7801037430763245, -0.2376929521560669], "gt_se_raw": [16.0, 52.0], "gt_se_norm": [-0.7654439806938171, -0.2376929521560669], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005030.pt"}
{"idx": 5032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ck2K69zZBG4.mp4", "duration": 136.42795833333332, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the man says 'Thank you to Dad's first 100 subscribers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the man says 'Thank you to Dad's first 100 subscribers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 106.0], "base_se_norm": [0.49529457092285156, 0.5539336204528809], "gt_se_raw": [101.0, 104.0], "gt_se_norm": [0.4806349277496338, 0.5246140956878662], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005032.pt"}
{"idx": 5034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EthCVn45VyU.mp4", "duration": 45.8, "prompt_text": "Please find the visual event described by the sentence 'When is the person actively washing their hands under the running water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person actively washing their hands under the running water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 44.0], "base_se_norm": [-0.7816593647003174, 0.9213974475860596], "gt_se_raw": [4.0, 34.0], "gt_se_norm": [-0.8253275156021118, 0.48471617698669434], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005034.pt"}
{"idx": 5036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EthCVn45VyU.mp4", "duration": 45.8, "prompt_text": "Please find the visual event described by the sentence 'At what point are the hands being rinsed to remove the soap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the hands being rinsed to remove the soap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 43.0], "base_se_norm": [0.1790393590927124, 0.8777292966842651], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [0.1790393590927124, 0.48471617698669434], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005036.pt"}
{"idx": 5038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EthCVn45VyU.mp4", "duration": 45.8, "prompt_text": "Please find the visual event described by the sentence 'Identify the entire duration when water is flowing from the faucet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the entire duration when water is flowing from the faucet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 44.0], "base_se_norm": [-0.8253275156021118, 0.9213974475860596], "gt_se_raw": [3.0, 34.0], "gt_se_norm": [-0.8689956665039062, 0.48471617698669434], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005038.pt"}
{"idx": 5040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVzDVbkDleE.mp4", "duration": 25.158466666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Plankton is revealed to be controlling SpongeBob's brain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Plankton is revealed to be controlling SpongeBob's brain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 15.0], "base_se_norm": [-0.682015597820282, 0.19244158267974854], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7615116834640503, -0.5230233669281006], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005040.pt"}
{"idx": 5042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVzDVbkDleE.mp4", "duration": 25.158466666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where SpongeBob, in his underwear, begs Plankton to leave his brain alone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where SpongeBob, in his underwear, begs Plankton to leave his brain alone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [0.27193760871887207, 0.7489142417907715], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.27193760871887207, 0.4309297800064087], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005042.pt"}
{"idx": 5044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ajzArdLK6tE.mp4", "duration": 143.56008333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the spices for the dill pickles added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the spices for the dill pickles added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 18.0], "base_se_norm": [-0.902479887008667, -0.7492339611053467], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.8328226208686829, -0.7770968079566956], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005044.pt"}
{"idx": 5046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ajzArdLK6tE.mp4", "duration": 143.56008333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point are the ingredients for the bread and butter pickle brine mixed in the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the ingredients for the bread and butter pickle brine mixed in the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 67.0], "base_se_norm": [-0.33129048347473145, -0.06659293174743652], "gt_se_raw": [48.0, 66.0], "gt_se_norm": [-0.33129048347473145, -0.08052438497543335], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005046.pt"}
{"idx": 5048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ajzArdLK6tE.mp4", "duration": 143.56008333333332, "prompt_text": "Please find the visual event described by the sentence 'When are all the ingredients for the spicy pickles being boiled together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the ingredients for the spicy pickles being boiled together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 100.0], "base_se_norm": [0.21203601360321045, 0.3931448459625244], "gt_se_raw": [98.0, 100.0], "gt_se_norm": [0.36528193950653076, 0.3931448459625244], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005048.pt"}
{"idx": 5050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksqJTtIDgpw.mp4", "duration": 56.08, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman run past a metal fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman run past a metal fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.8573466539382935, -0.7146933078765869], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.8216832876205444, -0.7146933078765869], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005050.pt"}
{"idx": 5052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksqJTtIDgpw.mp4", "duration": 56.08, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman is walking on a fallen tree trunk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman is walking on a fallen tree trunk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [0.24821674823760986, 0.3908700942993164], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [0.24821674823760986, 0.35520684719085693], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005052.pt"}
{"idx": 5054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I1Ksh6qrFNc.mp4", "duration": 29.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man stand in front of a white building with pillars?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man stand in front of a white building with pillars?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.3986636996269226, -0.2650333642959595], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.3986636996269226, -0.2650333642959595], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005054.pt"}
{"idx": 5056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I1Ksh6qrFNc.mp4", "duration": 29.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man pointing his finger while sitting inside the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man pointing his finger while sitting inside the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [0.20267260074615479, 0.33630287647247314], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.20267260074615479, 0.33630287647247314], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005056.pt"}
{"idx": 5058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I1Ksh6qrFNc.mp4", "duration": 29.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man is yelling out of the car window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man is yelling out of the car window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.5991091132164001, -0.3986636996269226], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [0.6703786849975586, 0.804008960723877], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005058.pt"}
{"idx": 5060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i8Ol7OPpnbo.mp4", "duration": 27.260566666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the girl in the white sweater playfully touches the other girl's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the girl in the white sweater playfully touches the other girl's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9266339540481567, -0.2663395404815674], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.0462414026260376, 0.10049068927764893], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005060.pt"}
{"idx": 5062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i8Ol7OPpnbo.mp4", "duration": 27.260566666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the two individuals are playfully interacting with their hands over the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the two individuals are playfully interacting with their hands over the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, 0.687419056892395], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.7799018621444702], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005062.pt"}
{"idx": 5064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkX3eBvPDFQ.mp4", "duration": 301.88, "prompt_text": "Please find the visual event described by the sentence 'When does the little ostrich try to take off by flapping its wings and running?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little ostrich try to take off by flapping its wings and running?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 45.0], "base_se_norm": [-0.8343712687492371, -0.7018682956695557], "gt_se_raw": [26.0, 32.0], "gt_se_norm": [-0.8277461528778076, -0.7879952192306519], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005064.pt"}
{"idx": 5066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkX3eBvPDFQ.mp4", "duration": 301.88, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the ostrich gets caught in the forked tree, pulls back, and launches itself like a slingshot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the ostrich gets caught in the forked tree, pulls back, and launches itself like a slingshot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 199.0], "base_se_norm": [0.1594010591506958, 0.3184046745300293], "gt_se_raw": [188.0, 198.0], "gt_se_norm": [0.245527982711792, 0.3117794990539551], "start_token_span": [80, 85], "end_token_span": [87, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005066.pt"}
{"idx": 5068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkX3eBvPDFQ.mp4", "duration": 301.88, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the ostrich, covered in flowers, uses a spiderweb as a parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 214.0 - 261.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the ostrich, covered in flowers, uses a spiderweb as a parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 214.0 - 261.0 seconds.<|im_end|>\n", "base_se_raw": [214.0, 261.0], "base_se_norm": [0.4177818298339844, 0.7291638851165771], "gt_se_raw": [234.0, 242.0], "gt_se_norm": [0.5502848625183105, 0.6032860279083252], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005068.pt"}
{"idx": 5070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h3gV_MPxr3I.mp4", "duration": 153.01953333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the man guides the boy towards the ball using a hand-under-hand technique.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the man guides the boy towards the ball using a hand-under-hand technique.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.5425420999526978, -0.46412068605422974], "gt_se_raw": [36.0, 42.0], "gt_se_norm": [-0.5294718742370605, -0.45105046033859253], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005070.pt"}
{"idx": 5072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h3gV_MPxr3I.mp4", "duration": 153.01953333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy push the ball, causing it to roll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy push the ball, causing it to roll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 27.0], "base_se_norm": [-0.7385954856872559, -0.6471039056777954], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.7124550342559814, -0.686314582824707], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005072.pt"}
{"idx": 5074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ha92yzmVv2s.mp4", "duration": 113.7, "prompt_text": "Please find the visual event described by the sentence 'When is the brown and white dog searching for the yellow toy for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the brown and white dog searching for the yellow toy for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 53.0], "base_se_norm": [-0.33157432079315186, -0.06772202253341675], "gt_se_raw": [42.0, 51.0], "gt_se_norm": [-0.2612137198448181, -0.102902352809906], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005074.pt"}
{"idx": 5076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ha92yzmVv2s.mp4", "duration": 113.7, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog successfully retrieve the toy without verbal guidance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog successfully retrieve the toy without verbal guidance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 55.0], "base_se_norm": [-0.1380826234817505, -0.032541751861572266], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [0.7766051292419434, 0.8469657897949219], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005076.pt"}
{"idx": 5078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ha92yzmVv2s.mp4", "duration": 113.7, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the brown and white dog jumps and barks excitedly before a throw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the brown and white dog jumps and barks excitedly before a throw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8065083622932434, -0.718557596206665], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [0.09058928489685059, 0.14335978031158447], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005078.pt"}
{"idx": 5080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g6KUxEfUVm0.mp4", "duration": 49.816433333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man massages a woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man massages a woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.47808384895324707, -0.19705212116241455], "gt_se_raw": [13.0, 20.0], "gt_se_norm": [-0.47808384895324707, -0.19705212116241455], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005080.pt"}
{"idx": 5082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g6KUxEfUVm0.mp4", "duration": 49.816433333333336, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a man in a beige suit jacket is observing the massage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a man in a beige suit jacket is observing the massage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 36.0], "base_se_norm": [0.1642744541168213, 0.4453061819076538], "gt_se_raw": [29.0, 36.0], "gt_se_norm": [0.1642744541168213, 0.4453061819076538], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005082.pt"}
{"idx": 5084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yZyxJxR6RCA.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When is a man in a black jacket adjusting the camera on a tripod?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man in a black jacket adjusting the camera on a tripod?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.934491753578186, -0.6069504022598267], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.934491753578186, -0.7379669547080994], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005084.pt"}
{"idx": 5086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yZyxJxR6RCA.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man with glasses looking thoughtfully at a computer monitor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man with glasses looking thoughtfully at a computer monitor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005086.pt"}
{"idx": 5088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yZyxJxR6RCA.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'YOU'VE GOT THIS!' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'YOU'VE GOT THIS!' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5066900253295898, 0.8342313766479492], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [0.5066900253295898, 0.8342313766479492], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005088.pt"}
{"idx": 5090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZ0Bey4YUGI.mp4", "duration": 142.64249999999998, "prompt_text": "Please find the visual event described by the sentence 'A giant hand reaches through a window and grabs a person from their bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A giant hand reaches through a window and grabs a person from their bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.4391573667526245, -0.39709413051605225], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.4391573667526245, -0.383073091506958], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005090.pt"}
{"idx": 5092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZ0Bey4YUGI.mp4", "duration": 142.64249999999998, "prompt_text": "Please find the visual event described by the sentence 'The giant holds the girl in the palm of his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The giant holds the girl in the palm of his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 113.0], "base_se_norm": [0.542317271232605, 0.5843805074691772], "gt_se_raw": [110.0, 113.0], "gt_se_norm": [0.542317271232605, 0.5843805074691772], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005092.pt"}
{"idx": 5094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a7CpzJ-sNl8.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a blue striped shirt working on two computer monitors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a blue striped shirt working on two computer monitors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.2335669994354248, -0.06695109605789185], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.2335669994354248, -0.06695109605789185], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005094.pt"}
{"idx": 5096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a7CpzJ-sNl8.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment where a student is receiving a diploma on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment where a student is receiving a diploma on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [0.566189169883728, 0.7328051328659058], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.6328355073928833, 0.7328051328659058], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005096.pt"}
{"idx": 5098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a7CpzJ-sNl8.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip of an airplane flying in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip of an airplane flying in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 3.0], "base_se_norm": [-0.9333536624908447, -0.9000304937362671], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9666768312454224, -0.9000304937362671], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005098.pt"}
{"idx": 5100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cqDG7q508iM.mp4", "duration": 330.5969333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when both men are laughing hysterically as the rules are being explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when both men are laughing hysterically as the rules are being explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [-0.7761624455451965, -0.7459141612052917], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.8064107894897461, -0.7519637942314148], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005100.pt"}
{"idx": 5102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cqDG7q508iM.mp4", "duration": 330.5969333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man with the beard gets a raw egg smashed on his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man with the beard gets a raw egg smashed on his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 82.0], "base_se_norm": [-0.5220766067504883, -0.5039275884628296], "gt_se_raw": [247.0, 250.0], "gt_se_norm": [0.4942667484283447, 0.5124157667160034], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005102.pt"}
{"idx": 5104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hlgcYUM35yY.mp4", "duration": 177.16, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is using a purple comb to part her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is using a purple comb to part her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [-0.7742154002189636, -0.7064800262451172], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.7629261612892151, -0.7290585041046143], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005104.pt"}
{"idx": 5106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hlgcYUM35yY.mp4", "duration": 177.16, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where she divides the hair into three sections and labels them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where she divides the hair into three sections and labels them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.5484308004379272, -0.4129600524902344], "gt_se_raw": [40.0, 51.0], "gt_se_norm": [-0.5484308004379272, -0.4242492914199829], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005106.pt"}
{"idx": 5108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hlgcYUM35yY.mp4", "duration": 177.16, "prompt_text": "Please find the visual event described by the sentence 'Show the final look with two completed braids.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 175.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the final look with two completed braids.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 175.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 175.0], "base_se_norm": [0.8175660371780396, 0.9756152629852295], "gt_se_raw": [161.0, 168.0], "gt_se_norm": [0.8175660371780396, 0.8965905904769897], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005108.pt"}
{"idx": 5110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOaoI6-YAw4.mp4", "duration": 147.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when a player is sitting on a wooden sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when a player is sitting on a wooden sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [-0.15665388107299805, -0.10224443674087524], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [-0.15665388107299805, -0.12944912910461426], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005110.pt"}
{"idx": 5112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOaoI6-YAw4.mp4", "duration": 147.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the 'VICTORY!' screen appearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the 'VICTORY!' screen appearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 106.0], "base_se_norm": [0.4146451950073242, 0.44184982776641846], "gt_se_raw": [92.0, 94.0], "gt_se_norm": [0.25141680240631104, 0.2786215543746948], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005112.pt"}
{"idx": 5114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r35hkbGnHgM.mp4", "duration": 26.292933333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a windmill silhouetted against a full moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a windmill silhouetted against a full moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.7718018293380737], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.7718018293380737], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005114.pt"}
{"idx": 5116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r35hkbGnHgM.mp4", "duration": 26.292933333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are people ice skating on a frozen canal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people ice skating on a frozen canal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.23933929204940796, 0.06492495536804199], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.23933929204940796, 0.06492495536804199], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005116.pt"}
{"idx": 5118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r35hkbGnHgM.mp4", "duration": 26.292933333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time range showing an aerial shot of a star-shaped fort.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range showing an aerial shot of a star-shaped fort.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [0.3691892623901367, 0.7495195865631104], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [0.29312312602996826, 0.6734534502029419], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005118.pt"}
{"idx": 5120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HTLgn4bZu3w.mp4", "duration": 61.89516666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman speaking while wearing glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman speaking while wearing glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [0.421758770942688, 0.6479476690292358], "gt_se_raw": [44.0, 52.0], "gt_se_norm": [0.421758770942688, 0.680260419845581], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005120.pt"}
{"idx": 5122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HTLgn4bZu3w.mp4", "duration": 61.89516666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman is sitting and talking by herself, before any photos are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman is sitting and talking by herself, before any photos are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.8707492351531982, -0.7091857194900513], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.8707492351531982, -0.7091857194900513], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005122.pt"}
{"idx": 5124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zywWM3J3i8M.mp4", "duration": 149.6495, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip that shows two women watching a movie and smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip that shows two women watching a movie and smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 40.0], "base_se_norm": [-0.5990631580352783, -0.4654175639152527], "gt_se_raw": [30.0, 39.0], "gt_se_norm": [-0.5990631580352783, -0.47878211736679077], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005124.pt"}
{"idx": 5126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zywWM3J3i8M.mp4", "duration": 149.6495, "prompt_text": "Please find the visual event described by the sentence 'Show me when the animated IMAX logo appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the animated IMAX logo appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 101.0], "base_se_norm": [0.29636240005493164, 0.34982073307037354], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [0.3097269535064697, 0.34982073307037354], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005126.pt"}
{"idx": 5128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v0jX8zWZaok.mp4", "duration": 309.684375, "prompt_text": "Please find the visual event described by the sentence 'When does the man put on his headphones at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man put on his headphones at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9806254506111145, -0.9483345150947571], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9741672277450562, -0.961250901222229], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005128.pt"}
{"idx": 5130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v0jX8zWZaok.mp4", "duration": 309.684375, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man riding the ATV with his arms raised in joy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man riding the ATV with his arms raised in joy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 121.0], "base_se_norm": [-0.23147565126419067, -0.21855926513671875], "gt_se_raw": [119.0, 121.0], "gt_se_norm": [-0.23147565126419067, -0.21855926513671875], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005130.pt"}
{"idx": 5132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v0jX8zWZaok.mp4", "duration": 309.684375, "prompt_text": "Please find the visual event described by the sentence 'When does the man get attacked by another person for the notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 209.0 - 221.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get attacked by another person for the notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 209.0 - 221.0 seconds.<|im_end|>\n", "base_se_raw": [209.0, 221.0], "base_se_norm": [0.3497612476348877, 0.42725956439971924], "gt_se_raw": [211.0, 221.0], "gt_se_norm": [0.3626776933670044, 0.42725956439971924], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005132.pt"}
{"idx": 5134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C7uAB94aRrQ.mp4", "duration": 29.988291666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the bulldog is driving the truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the bulldog is driving the truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.5331511497497559, -0.26638031005859375], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [-0.3997657299041748, -0.3330730199813843], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005134.pt"}
{"idx": 5136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C7uAB94aRrQ.mp4", "duration": 29.988291666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the truck honk its horn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the truck honk its horn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [0.13377583026885986, 0.2671612501144409], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [0.0003904104232788086, 0.06708312034606934], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005136.pt"}
{"idx": 5138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wAnrYQGSHMw.mp4", "duration": 175.71720833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the introduction of the trainers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the introduction of the trainers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 25.0], "base_se_norm": [-0.8520349860191345, -0.7154518961906433], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [-0.8406530618667603, -0.7837434411048889], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005138.pt"}
{"idx": 5140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wAnrYQGSHMw.mp4", "duration": 175.71720833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the trainers are doing tricep dips on the chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the trainers are doing tricep dips on the chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 58.0], "base_se_norm": [-0.5105772614479065, -0.3398483991622925], "gt_se_raw": [44.0, 57.0], "gt_se_norm": [-0.4991953372955322, -0.35123032331466675], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005140.pt"}
{"idx": 5142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wAnrYQGSHMw.mp4", "duration": 175.71720833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the trainers performing the bird-dog exercise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the trainers performing the bird-dog exercise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 58.0], "base_se_norm": [-0.5105772614479065, -0.3398483991622925], "gt_se_raw": [137.0, 147.0], "gt_se_norm": [0.5593236684799194, 0.6731429100036621], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005142.pt"}
{"idx": 5144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AoKlbyqbEGM.mp4", "duration": 32.76606666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman in the white cardigan is looking up and to her right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman in the white cardigan is looking up and to her right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.6337674856185913, -0.4506511688232422], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.57272869348526, -0.4506511688232422], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005144.pt"}
{"idx": 5146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AoKlbyqbEGM.mp4", "duration": 32.76606666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman help the girl pluck individual strings on the ukulele?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman help the girl pluck individual strings on the ukulele?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.20649617910385132], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.26753491163253784, -0.023379921913146973], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005146.pt"}
{"idx": 5148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1MsvTOTtr8A.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'At what point is the mascot reading a magazine while sitting in a chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the mascot reading a magazine while sitting in a chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.5337995290756226, -0.30069929361343384], "gt_se_raw": [13.0, 21.0], "gt_se_norm": [-0.5670995712280273, -0.30069929361343384], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005148.pt"}
{"idx": 5150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1MsvTOTtr8A.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'When can we see the mascot cleaning a glass trophy case?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the mascot cleaning a glass trophy case?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.03429907560348511, 0.13220107555389404], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.06759911775588989, 0.09890103340148926], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005150.pt"}
{"idx": 5152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1MsvTOTtr8A.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the mascot is having a video call on a smartphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the mascot is having a video call on a smartphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 51.0], "base_se_norm": [0.49850142002105713, 0.6983016729354858], "gt_se_raw": [43.0, 51.0], "gt_se_norm": [0.4319014549255371, 0.6983016729354858], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005152.pt"}
{"idx": 5154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3mG0XW9RoNM.mp4", "duration": 95.01158333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point is a hair straightener used on the doll's blonde and pink hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a hair straightener used on the doll's blonde and pink hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.894749641418457, -0.8105493783950806], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.894749641418457, -0.8105493783950806], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005154.pt"}
{"idx": 5156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3mG0XW9RoNM.mp4", "duration": 95.01158333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the doll's original makeup is being removed from its face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the doll's original makeup is being removed from its face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.2632477283477783, -0.20009750127792358], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.2632477283477783, -0.2211475968360901], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005156.pt"}
{"idx": 5158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsVV0t_cS7Q.mp4", "duration": 119.54166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where they demonstrate how thick the frosty is by turning the cups upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where they demonstrate how thick the frosty is by turning the cups upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 44.0], "base_se_norm": [-0.46462178230285645, -0.26385498046875], "gt_se_raw": [32.0, 44.0], "gt_se_norm": [-0.46462178230285645, -0.26385498046875], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005158.pt"}
{"idx": 5160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsVV0t_cS7Q.mp4", "duration": 119.54166666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the three men start drinking their milkshakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the three men start drinking their milkshakes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 60.0], "base_se_norm": [-0.0798187255859375, 0.0038341283798217773], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.09654927253723145, -0.06308817863464355], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005160.pt"}
{"idx": 5162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsVV0t_cS7Q.mp4", "duration": 119.54166666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the men are reacting to the brain freeze.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the men are reacting to the brain freeze.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 96.0], "base_se_norm": [0.17113983631134033, 0.6061345338821411], "gt_se_raw": [97.0, 103.0], "gt_se_norm": [0.6228652000427246, 0.7232486009597778], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005162.pt"}
{"idx": 5164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pNIGmCzL42g.mp4", "duration": 161.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman in the vest holds the child's hands to calm them down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman in the vest holds the child's hands to calm them down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 78.0], "base_se_norm": [-0.5778145790100098, -0.031456947326660156], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.5902317762374878, -0.5281456708908081], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005164.pt"}
{"idx": 5166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pNIGmCzL42g.mp4", "duration": 161.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman presents the open yellow box to the child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman presents the open yellow box to the child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 66.0], "base_se_norm": [-0.30463576316833496, -0.18046355247497559], "gt_se_raw": [89.0, 94.0], "gt_se_norm": [0.10513246059417725, 0.16721856594085693], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005166.pt"}
{"idx": 5168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ioq5yWKwE5o.mp4", "duration": 369.6, "prompt_text": "Please find the visual event described by the sentence 'When does the woman place the Santa Claus picture on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman place the Santa Claus picture on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 66.0], "base_se_norm": [-0.6536796689033508, -0.6428571343421936], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [-0.6536796689033508, -0.6428571343421936], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005168.pt"}
{"idx": 5170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ioq5yWKwE5o.mp4", "duration": 369.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman first place the Christmas tree card on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman first place the Christmas tree card on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 131.0], "base_se_norm": [-0.3235930800437927, -0.2911255359649658], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.8863636255264282, -0.8755411505699158], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005170.pt"}
{"idx": 5172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ioq5yWKwE5o.mp4", "duration": 369.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the girl claps her hands excitedly in response to the Santa card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the girl claps her hands excitedly in response to the Santa card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 113.0], "base_se_norm": [-0.4318181872367859, -0.38852816820144653], "gt_se_raw": [106.0, 109.0], "gt_se_norm": [-0.4264069199562073, -0.4101731777191162], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005172.pt"}
{"idx": 5174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RRt2YsawMj8.mp4", "duration": 99.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the student touches her hearing aid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the student touches her hearing aid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 25.0], "base_se_norm": [-0.5978552103042603, -0.4973190426826477], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [-0.5777480006217957, -0.4973190426826477], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005174.pt"}
{"idx": 5176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RRt2YsawMj8.mp4", "duration": 99.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the student says 'I am ready'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the student says 'I am ready'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.25603216886520386, -0.19571048021316528], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [0.18632709980010986, 0.22654151916503906], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005176.pt"}
{"idx": 5178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10fmOuVnbco.mp4", "duration": 115.44866666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the green hoodie first touch the small, light green paper object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the green hoodie first touch the small, light green paper object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9133813977241516, -0.826762855052948], "gt_se_raw": [1.0, 9.0], "gt_se_norm": [-0.9826762676239014, -0.8440865278244019], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005178.pt"}
{"idx": 5180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10fmOuVnbco.mp4", "duration": 115.44866666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the blindfolded person compare both paper spheres by holding one in each hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the blindfolded person compare both paper spheres by holding one in each hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 74.0], "base_se_norm": [0.07407045364379883, 0.2819550037384033], "gt_se_raw": [79.0, 90.0], "gt_se_norm": [0.36857354640960693, 0.5591344833374023], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005180.pt"}
{"idx": 5182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10fmOuVnbco.mp4", "duration": 115.44866666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the woman in green opens the second sphere, revealing the pictures inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the woman in green opens the second sphere, revealing the pictures inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 77.0], "base_se_norm": [0.07407045364379883, 0.3339262008666992], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [0.3166024684906006, 0.4032210111618042], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005182.pt"}
{"idx": 5184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OUk4LxF5nzw.mp4", "duration": 405.0, "prompt_text": "Please find the visual event described by the sentence 'When is the car passing a yellow bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the car passing a yellow bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 118.0], "base_se_norm": [-0.44197529554367065, -0.41728395223617554], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [-0.43209874629974365, -0.41728395223617554], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005184.pt"}
{"idx": 5186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OUk4LxF5nzw.mp4", "duration": 405.0, "prompt_text": "Please find the visual event described by the sentence 'When is the passenger seen filming with a camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the passenger seen filming with a camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 173.0], "base_se_norm": [-0.3629629611968994, -0.14567899703979492], "gt_se_raw": [130.0, 142.0], "gt_se_norm": [-0.3580247163772583, -0.2987654209136963], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005186.pt"}
{"idx": 5188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JCOqo88eW1E.mp4", "duration": 33.9339, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the girl's hand to hold the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the girl's hand to hold the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.23380452394485474], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5874332189559937], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005188.pt"}
{"idx": 5190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JCOqo88eW1E.mp4", "duration": 33.9339, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the girl raises the spoon high into the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the girl raises the spoon high into the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [0.29663848876953125, 0.5323909521102905], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [0.23770034313201904, 0.35557663440704346], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005190.pt"}
{"idx": 5192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JCOqo88eW1E.mp4", "duration": 33.9339, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the girl successfully takes a bite of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the girl successfully takes a bite of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [0.06088602542877197, 0.1787623167037964], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [0.7681434154510498, 0.8860195875167847], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005192.pt"}
{"idx": 5194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k0jstVk5AiU.mp4", "duration": 171.08758333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the animation explaining the 6-month grace period for Stafford loans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the animation explaining the 6-month grace period for Stafford loans.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 64.0], "base_se_norm": [-0.4038141369819641, -0.25184518098831177], "gt_se_raw": [51.0, 64.0], "gt_se_norm": [-0.4038141369819641, -0.25184518098831177], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005194.pt"}
{"idx": 5196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k0jstVk5AiU.mp4", "duration": 171.08758333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the presenter in the music studio shown speaking for the final time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the presenter in the music studio shown speaking for the final time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 157.0], "base_se_norm": [0.6833482980728149, 0.8353172540664673], "gt_se_raw": [144.0, 157.0], "gt_se_norm": [0.6833482980728149, 0.8353172540664673], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005196.pt"}
{"idx": 5198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vvDIVVVAx-k.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the bar chart with the title 'Employment for disabled and non-disabled Americans'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the bar chart with the title 'Employment for disabled and non-disabled Americans'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [82, 85], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005198.pt"}
{"idx": 5200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vvDIVVVAx-k.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the narrator provides a summary stating people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the narrator provides a summary stating people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.030096113681793213, 0.2608751058578491], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.12708652019500732, 0.2608751058578491], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005200.pt"}
{"idx": 5202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vvDIVVVAx-k.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the narrator begins to talk about the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the narrator begins to talk about the chart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.4180576801300049], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005202.pt"}
{"idx": 5204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GT018Qejkhg.mp4", "duration": 122.6225, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman wearing a denim shirt and a scarf introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman wearing a denim shirt and a scarf introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 32.0], "base_se_norm": [-0.6737955808639526, -0.4780729413032532], "gt_se_raw": [21.0, 32.0], "gt_se_norm": [-0.6574853658676147, -0.4780729413032532], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005204.pt"}
{"idx": 5206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GT018Qejkhg.mp4", "duration": 122.6225, "prompt_text": "Please find the visual event described by the sentence 'Show when the speaker explains that their development calendar is about two years ahead.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show when the speaker explains that their development calendar is about two years ahead.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 63.0], "base_se_norm": [-0.08662766218185425, 0.02754390239715576], "gt_se_raw": [49.0, 60.0], "gt_se_norm": [-0.20079916715621948, -0.02138674259185791], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005206.pt"}
{"idx": 5208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IANwb_qT1gg.mp4", "duration": 160.64, "prompt_text": "Please find the visual event described by the sentence 'When does the man demonstrate the throttle control of the model airplane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man demonstrate the throttle control of the model airplane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 58.0], "base_se_norm": [-0.6389442086219788, -0.2778884172439575], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.3152390122413635, -0.2778884172439575], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005208.pt"}
{"idx": 5210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IANwb_qT1gg.mp4", "duration": 160.64, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the man is chasing the plane as it takes off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the man is chasing the plane as it takes off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 100.0], "base_se_norm": [0.10806775093078613, 0.24501991271972656], "gt_se_raw": [92.0, 96.0], "gt_se_norm": [0.14541828632354736, 0.19521915912628174], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005210.pt"}
{"idx": 5212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IANwb_qT1gg.mp4", "duration": 160.64, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the squirrel jumps out of the plane just before it crashes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the squirrel jumps out of the plane just before it crashes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 142.0], "base_se_norm": [0.7181274890899658, 0.7679282426834106], "gt_se_raw": [139.0, 142.0], "gt_se_norm": [0.7305777072906494, 0.7679282426834106], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005212.pt"}
{"idx": 5214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ucYIwvZYGp8.mp4", "duration": 165.24841666666666, "prompt_text": "Please find the visual event described by the sentence 'When do the men get off the inflatable boat and walk on the shore?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the men get off the inflatable boat and walk on the shore?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.6127042770385742, -0.5884983539581299], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.6127042770385742, -0.5884983539581299], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005214.pt"}
{"idx": 5216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ucYIwvZYGp8.mp4", "duration": 165.24841666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a white chef's uniform addresses his staff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a white chef's uniform addresses his staff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [-0.7942491769790649, -0.7700431942939758], "gt_se_raw": [80.0, 82.0], "gt_se_norm": [-0.03176075220108032, -0.007554769515991211], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005216.pt"}
{"idx": 5218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/URdX9rFIbcc.mp4", "duration": 193.56837499999997, "prompt_text": "Please find the visual event described by the sentence 'When are the fish fillets fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the fish fillets fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 158.0], "base_se_norm": [0.5601722002029419, 0.632498025894165], "gt_se_raw": [152.0, 157.0], "gt_se_norm": [0.5705044269561768, 0.6221657991409302], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005218.pt"}
{"idx": 5220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/URdX9rFIbcc.mp4", "duration": 193.56837499999997, "prompt_text": "Please find the visual event described by the sentence 'At what point is the cheddar cheese added to the macaroni?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the cheddar cheese added to the macaroni?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 122.0], "base_se_norm": [0.20887517929077148, 0.2605365514755249], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.20887517929077148, 0.23987197875976562], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005220.pt"}
{"idx": 5222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/URdX9rFIbcc.mp4", "duration": 193.56837499999997, "prompt_text": "Please find the visual event described by the sentence 'When are the chicken pieces cooked in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the chicken pieces cooked in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.9173418879508972, -0.8346837759017944], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.9070096015930176, -0.8346837759017944], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005222.pt"}
{"idx": 5224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4v22j4oiTsg.mp4", "duration": 430.13013013013017, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of stirring the yogurt and cream mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of stirring the yogurt and cream mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 51.0], "base_se_norm": [-0.800060510635376, -0.7628624439239502], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [-0.7954107522964478, -0.7628624439239502], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005224.pt"}
{"idx": 5226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4v22j4oiTsg.mp4", "duration": 430.13013013013017, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when flour and baking powder are sifted into the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 205.0 - 243.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when flour and baking powder are sifted into the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 205.0 - 243.0 seconds.<|im_end|>\n", "base_se_raw": [205.0, 243.0], "base_se_norm": [-0.046800076961517334, 0.12989068031311035], "gt_se_raw": [237.0, 243.0], "gt_se_norm": [0.10199213027954102, 0.12989068031311035], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005226.pt"}
{"idx": 5228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-BIL6p1Te8.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the finished red origami heart is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the finished red origami heart is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 138.0], "base_se_norm": [0.561684250831604, 0.6451330184936523], "gt_se_raw": [131.0, 135.0], "gt_se_norm": [0.561684250831604, 0.609369158744812], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005228.pt"}
{"idx": 5230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-BIL6p1Te8.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'When is the paper model flipped over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the paper model flipped over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 100.0], "base_se_norm": [0.13251912593841553, 0.1921253204345703], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.14444029331207275, 0.1682828664779663], "start_token_span": [62, 66], "end_token_span": [68, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005230.pt"}
{"idx": 5232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-BIL6p1Te8.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'At what time is the bottom corner folded up to the top edge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the bottom corner folded up to the top edge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 60.0], "base_se_norm": [-0.3681735396385193, -0.28472477197647095], "gt_se_raw": [59.0, 68.0], "gt_se_norm": [-0.2966460585594177, -0.1893547773361206], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005232.pt"}
{"idx": 5234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/akYVGzeS7_A.mp4", "duration": 85.11913333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the process of an organ being removed from the left side of the torso.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of an organ being removed from the left side of the torso.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 30.0], "base_se_norm": [-0.5535668730735779, -0.29510557651519775], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.5065739154815674, -0.4595809578895569], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005234.pt"}
{"idx": 5236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/akYVGzeS7_A.mp4", "duration": 85.11913333333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the brain is taken out of the skull.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the brain is taken out of the skull.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [0.4567817449569702, 0.5272711515426636], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.4567817449569702, 0.5037747621536255], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005236.pt"}
{"idx": 5238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R4WDWpR4oRM.mp4", "duration": 198.86533333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the flashing cube?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the flashing cube?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 25.0], "base_se_norm": [-0.9597717523574829, -0.7485735416412354], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.949714720249176, -0.9195435643196106], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005238.pt"}
{"idx": 5240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R4WDWpR4oRM.mp4", "duration": 198.86533333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time do the man and woman hug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do the man and woman hug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 104.0], "base_se_norm": [0.005705714225769043, 0.04593396186828613], "gt_se_raw": [101.0, 104.0], "gt_se_norm": [0.015762805938720703, 0.04593396186828613], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005240.pt"}
{"idx": 5242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R4WDWpR4oRM.mp4", "duration": 198.86533333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the leather jacket appears and sips from his mug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the leather jacket appears and sips from his mug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 158.0], "base_se_norm": [0.4985015392303467, 0.5890151262283325], "gt_se_raw": [149.0, 155.0], "gt_se_norm": [0.4985015392303467, 0.5588438510894775], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005242.pt"}
{"idx": 5244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DKoJm4OaZNw.mp4", "duration": 84.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of cutting four equal size circles from the calendar page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of cutting four equal size circles from the calendar page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.8334126472473145, -0.643027126789093], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.7382198572158813, -0.643027126789093], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005244.pt"}
{"idx": 5246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DKoJm4OaZNw.mp4", "duration": 84.04, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a red ribbon is tied around the completed envelope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a red ribbon is tied around the completed envelope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 62.0], "base_se_norm": [0.30890047550201416, 0.4754878282546997], "gt_se_raw": [55.0, 63.0], "gt_se_norm": [0.30890047550201416, 0.49928605556488037], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005246.pt"}
{"idx": 5248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10UKOhzoOws.mp4", "duration": 145.56, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the speaker discusses the challenges people with disabilities face in the digital space.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the speaker discusses the challenges people with disabilities face in the digital space.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 34.0], "base_se_norm": [-0.8625996112823486, -0.5328387022018433], "gt_se_raw": [17.0, 26.0], "gt_se_norm": [-0.7664193511009216, -0.6427589654922485], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005248.pt"}
{"idx": 5250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10UKOhzoOws.mp4", "duration": 145.56, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that showcases the Australian Chamber Orchestra website's accessibility features.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that showcases the Australian Chamber Orchestra website's accessibility features.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 78.0], "base_se_norm": [-0.38169825077056885, 0.07172298431396484], "gt_se_raw": [51.0, 71.0], "gt_se_norm": [-0.2992580533027649, -0.024457275867462158], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005250.pt"}
{"idx": 5252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10UKOhzoOws.mp4", "duration": 145.56, "prompt_text": "Please find the visual event described by the sentence 'Show the slide detailing Google's innovations in accessibility.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the slide detailing Google's innovations in accessibility.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 106.0], "base_se_norm": [0.2091233730316162, 0.4564441442489624], "gt_se_raw": [88.0, 107.0], "gt_se_norm": [0.2091233730316162, 0.4701840877532959], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005252.pt"}
{"idx": 5254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qw6RD5S3e8o.mp4", "duration": 30.3, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the red puppet looks at the goldfish in the fishbowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the red puppet looks at the goldfish in the fishbowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.20792078971862793, -0.07590758800506592], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.14191418886184692, -0.009900987148284912], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005254.pt"}
{"idx": 5256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qw6RD5S3e8o.mp4", "duration": 30.3, "prompt_text": "Please find the visual event described by the sentence 'When is the name 'ELMO' being written in blue crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the name 'ELMO' being written in blue crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [0.45214521884918213, 0.6501650810241699], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [0.5181518793106079, 0.7161716222763062], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005256.pt"}
{"idx": 5258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bxBHjpuXvCY.mp4", "duration": 71.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a man sitting on the stairs and making gestures?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a man sitting on the stairs and making gestures?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.6382189393043518, -0.5269017219543457], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.6382189393043518, -0.5269017219543457], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005258.pt"}
{"idx": 5260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bxBHjpuXvCY.mp4", "duration": 71.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a woman wearing a red and white scarf while speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a woman wearing a red and white scarf while speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.44341373443603516, -0.3599258065223694], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.47124308347702026, -0.3877550959587097], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005260.pt"}
{"idx": 5262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bxBHjpuXvCY.mp4", "duration": 71.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a girl in a light green outfit is playing with toys on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a girl in a light green outfit is playing with toys on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 33.0], "base_se_norm": [-0.1094619631767273, -0.08163267374038696], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [-0.1372913122177124, -0.08163267374038696], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005262.pt"}
{"idx": 5264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EFeXMvjKWzo.mp4", "duration": 88.2, "prompt_text": "Please find the visual event described by the sentence 'When are the website, email, and phone number displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the website, email, and phone number displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 71.0], "base_se_norm": [0.4512472152709961, 0.6099773645401001], "gt_se_raw": [63.0, 71.0], "gt_se_norm": [0.4285714626312256, 0.6099773645401001], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005264.pt"}
{"idx": 5266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EFeXMvjKWzo.mp4", "duration": 88.2, "prompt_text": "Please find the visual event described by the sentence 'Find the scene showing a group of four engineers wearing blue hard hats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene showing a group of four engineers wearing blue hard hats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.8185940980911255, -0.6598639488220215], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.841269850730896, -0.6598639488220215], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005266.pt"}
{"idx": 5268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5jjeIH8Y6XM.mp4", "duration": 30.4304, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man cleaning the ceiling fan with a duster?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man cleaning the ceiling fan with a duster?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [0.11730372905731201, 0.38019871711730957], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [0.11730372905731201, 0.24875116348266602], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005268.pt"}
{"idx": 5270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5jjeIH8Y6XM.mp4", "duration": 30.4304, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the family open the large green Swiffer box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the family open the large green Swiffer box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.14559125900268555, 0.11730372905731201], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.07986754179000854, 0.11730372905731201], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005270.pt"}
{"idx": 5272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5jjeIH8Y6XM.mp4", "duration": 30.4304, "prompt_text": "Please find the visual event described by the sentence 'Where in the video does someone clean under the refrigerator with a mop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video does someone clean under the refrigerator with a mop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.6056575179100037, -0.4742100238800049], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.5116461515426636, 0.6430937051773071], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005272.pt"}
{"idx": 5274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KopKofh_-5E.mp4", "duration": 107.6075, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman talks about her work being a form of activism.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman talks about her work being a form of activism.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 46.0], "base_se_norm": [-0.40524590015411377, -0.14504098892211914], "gt_se_raw": [46.0, 58.0], "gt_se_norm": [-0.14504098892211914, 0.07799184322357178], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005274.pt"}
{"idx": 5276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KopKofh_-5E.mp4", "duration": 107.6075, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment where a woman in a white polka-dot shirt smiles broadly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment where a woman in a white polka-dot shirt smiles broadly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [-0.31231558322906494, -0.2193852663040161], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.2565574049949646, -0.2193852663040161], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005276.pt"}
{"idx": 5278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DfPMxdHZKsw.mp4", "duration": 99.91666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the 'BADLANDSCHUGS' logo animation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the 'BADLANDSCHUGS' logo animation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.7798165082931519, -0.6797330975532532], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.7597998380661011, -0.6797330975532532], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005278.pt"}
{"idx": 5280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DfPMxdHZKsw.mp4", "duration": 99.91666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man starts chugging the V8 juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man starts chugging the V8 juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 83.0], "base_se_norm": [0.401167631149292, 0.6613845825195312], "gt_se_raw": [70.0, 82.0], "gt_se_norm": [0.401167631149292, 0.6413677930831909], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005280.pt"}
{"idx": 5282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DfPMxdHZKsw.mp4", "duration": 99.91666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the text '12 Seconds!' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the text '12 Seconds!' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 88.0], "base_se_norm": [0.6613845825195312, 0.7614679336547852], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [0.6613845825195312, 0.7614679336547852], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005282.pt"}
{"idx": 5284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkfLc-sQZNs.mp4", "duration": 94.9949, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman first say they are 'waiting for the food'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman first say they are 'waiting for the food'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8947311639785767, -0.7894623875617981], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.7473548650741577, -0.7052473425865173], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005284.pt"}
{"idx": 5286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkfLc-sQZNs.mp4", "duration": 94.9949, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the man expresses agreement by repeatedly saying 'yeah'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the man expresses agreement by repeatedly saying 'yeah'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 44.0], "base_se_norm": [-0.17890334129333496, -0.07363450527191162], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [0.07374179363250732, 0.1369030475616455], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005286.pt"}
{"idx": 5288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aPH3XQbRdx4.mp4", "duration": 34.46776666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman and the man walking together down the hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman and the man walking together down the hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 31.0], "base_se_norm": [-0.9419747591018677, 0.7987821102142334], "gt_se_raw": [0.0, 15.0], "gt_se_norm": [-1.0, -0.12962156534194946], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005288.pt"}
{"idx": 5290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aPH3XQbRdx4.mp4", "duration": 34.46776666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment they stop walking and stand next to a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment they stop walking and stand next to a door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 31.0], "base_se_norm": [-0.1876468062400818, 0.7987821102142334], "gt_se_raw": [15.0, 26.0], "gt_se_norm": [-0.12962156534194946, 0.5086559057235718], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005290.pt"}
{"idx": 5292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aPH3XQbRdx4.mp4", "duration": 34.46776666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man turn and start walking back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man turn and start walking back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 20.0], "base_se_norm": [-0.4197477102279663, 0.16050457954406738], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [0.5086559057235718, 0.7987821102142334], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005292.pt"}
{"idx": 5294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99k-EAMBuM8.mp4", "duration": 51.44, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the duckling gets very close to the camera and appears to peck at the lens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the duckling gets very close to the camera and appears to peck at the lens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.2612752318382263, -0.06687402725219727], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.14463448524475098, -0.027993738651275635], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005294.pt"}
{"idx": 5296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99k-EAMBuM8.mp4", "duration": 51.44, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the person in khaki shorts is running quickly, chasing the duckling down a curved path.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the person in khaki shorts is running quickly, chasing the duckling down a curved path.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 48.0], "base_se_norm": [0.3996889591217041, 0.8662519454956055], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [0.36080873012542725, 0.6718508005142212], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005296.pt"}
{"idx": 5298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uOHvZjiDANg.mp4", "duration": 42.84, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man drawing at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man drawing at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 25.0], "base_se_norm": [-0.7198879718780518, 0.16713356971740723], "gt_se_raw": [6.0, 22.0], "gt_se_norm": [-0.7198879718780518, 0.027077436447143555], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005298.pt"}
{"idx": 5300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uOHvZjiDANg.mp4", "duration": 42.84, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cat is lying on the book under the lamp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cat is lying on the book under the lamp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [0.40056025981903076, 0.5873016119003296], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [0.44724559783935547, 0.5406162738800049], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005300.pt"}
{"idx": 5302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uOHvZjiDANg.mp4", "duration": 42.84, "prompt_text": "Please find the visual event described by the sentence 'Show the man's shocked reaction when he returns to the desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man's shocked reaction when he returns to the desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 37.0], "base_se_norm": [0.5873016119003296, 0.7273576259613037], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [0.5873016119003296, 0.7273576259613037], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005302.pt"}
{"idx": 5304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XVRLTfN1WYw.mp4", "duration": 89.1307, "prompt_text": "Please find the visual event described by the sentence 'When are the banana slices placed on the crust?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the banana slices placed on the crust?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.4390260577201843, -0.349270224571228], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.41658711433410645, -0.349270224571228], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005304.pt"}
{"idx": 5306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XVRLTfN1WYw.mp4", "duration": 89.1307, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the strawberry ice cream is being scooped.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the strawberry ice cream is being scooped.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.16975855827331543, -0.10244166851043701], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.16975855827331543, -0.12488067150115967], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005306.pt"}
{"idx": 5308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/__n8LfdLDmU.mp4", "duration": 197.76423333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the man turn to write on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man turn to write on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 81.0], "base_se_norm": [-0.26174718141555786, -0.1808428168296814], "gt_se_raw": [73.0, 81.0], "gt_se_norm": [-0.26174718141555786, -0.1808428168296814], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005308.pt"}
{"idx": 5310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/__n8LfdLDmU.mp4", "duration": 197.76423333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the man is holding a book and pen while pointing towards the audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the man is holding a book and pen while pointing towards the audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 32.0], "base_se_norm": [-1.0, -0.676382303237915], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.9898869395256042, -0.949434757232666], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005310.pt"}
{"idx": 5312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/__n8LfdLDmU.mp4", "duration": 197.76423333333335, "prompt_text": "Please find the visual event described by the sentence 'At what time does the 'BEYOND TODAY' logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 197.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the 'BEYOND TODAY' logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 197.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 197.0], "base_se_norm": [0.9417059421539307, 0.9922711849212646], "gt_se_raw": [192.0, 197.0], "gt_se_norm": [0.9417059421539307, 0.9922711849212646], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005312.pt"}
{"idx": 5314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U_g910Y1ptw.mp4", "duration": 27.92, "prompt_text": "Please find the visual event described by the sentence 'Show the girl complaining to her mother about the soup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the girl complaining to her mother about the soup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.5702005624771118, -0.2836676239967346], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.6418337821960449, -0.2836676239967346], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005314.pt"}
{"idx": 5316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U_g910Y1ptw.mp4", "duration": 27.92, "prompt_text": "Please find the visual event described by the sentence 'When does the girl start eating the soup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl start eating the soup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [0.43266475200653076, 0.862464189529419], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.36103153228759766, 0.575931191444397], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005316.pt"}
{"idx": 5318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DH2DgD9Rdi0.mp4", "duration": 60.166666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me the initial shot of the messy room before the dog is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the initial shot of the messy room before the dog is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9335179924964905], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9335179924964905], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005318.pt"}
{"idx": 5320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DH2DgD9Rdi0.mp4", "duration": 60.166666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the camera pans across the room to reveal more of the mess.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the camera pans across the room to reveal more of the mess.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 25.0], "base_se_norm": [-0.40166205167770386, -0.1689751148223877], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.3684210777282715, -0.20221608877182007], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005320.pt"}
{"idx": 5322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DH2DgD9Rdi0.mp4", "duration": 60.166666666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the dog turns back to look at the person after facing away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the dog turns back to look at the person after facing away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [0.6620497703552246, 0.761772871017456], "gt_se_raw": [51.0, 58.0], "gt_se_norm": [0.6952908039093018, 0.9279778003692627], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005322.pt"}
{"idx": 5324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghLkwSlWSXw.mp4", "duration": 197.03016666666664, "prompt_text": "Please find the visual event described by the sentence 'When is the boy speaking into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy speaking into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 58.0], "base_se_norm": [-0.47216206789016724, -0.41125768423080444], "gt_se_raw": [52.0, 59.0], "gt_se_norm": [-0.47216206789016724, -0.40110695362091064], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005324.pt"}
{"idx": 5326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghLkwSlWSXw.mp4", "duration": 197.03016666666664, "prompt_text": "Please find the visual event described by the sentence 'At what moment is an image of an electron microscope displayed on the laptop screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is an image of an electron microscope displayed on the laptop screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 137.0], "base_se_norm": [0.2789919376373291, 0.39065003395080566], "gt_se_raw": [126.0, 134.0], "gt_se_norm": [0.2789919376373291, 0.3601977825164795], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005326.pt"}
{"idx": 5328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SDeQT9zCvi4.mp4", "duration": 229.5293, "prompt_text": "Please find the visual event described by the sentence 'When can we see the yellow bird land on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the yellow bird land on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.8954381942749023, -0.8605842590332031], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.8954381942749023, -0.8780112266540527], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005328.pt"}
{"idx": 5330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SDeQT9zCvi4.mp4", "duration": 229.5293, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the donkey is wearing a hat and listening to a record player.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the donkey is wearing a hat and listening to a record player.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 68.0], "base_se_norm": [-0.5294718742370605, -0.4074830412864685], "gt_se_raw": [54.0, 61.0], "gt_se_norm": [-0.5294718742370605, -0.46847742795944214], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005330.pt"}
{"idx": 5332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SDeQT9zCvi4.mp4", "duration": 229.5293, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the donkey and the bird are both wearing sunglasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 172.0 - 196.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the donkey and the bird are both wearing sunglasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 172.0 - 196.0 seconds.<|im_end|>\n", "base_se_raw": [172.0, 196.0], "base_se_norm": [0.49871933460235596, 0.7078429460525513], "gt_se_raw": [171.0, 182.0], "gt_se_norm": [0.49000585079193115, 0.585854172706604], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005332.pt"}
{"idx": 5334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vcgn4EY5_S8.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when Furiosa kicks the man in the metal mask while he's on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when Furiosa kicks the man in the metal mask while he's on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.36063939332962036, -0.32970255613327026], "gt_se_raw": [90.0, 91.0], "gt_se_norm": [-0.07189589738845825, -0.06158357858657837], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005334.pt"}
{"idx": 5336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vcgn4EY5_S8.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the man in the mask finally gets the pistol and points it at the woman's head, ending the fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the man in the mask finally gets the pistol and points it at the woman's head, ending the fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 173.0], "base_se_norm": [0.7118364572525024, 0.784022331237793], "gt_se_raw": [164.0, 173.0], "gt_se_norm": [0.6912119388580322, 0.784022331237793], "start_token_span": [82, 87], "end_token_span": [89, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005336.pt"}
{"idx": 5338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0WWzgGyAH6Y.mp4", "duration": 219.17729166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the blue shirt is running away from a collapsing building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the blue shirt is running away from a collapsing building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 39.0], "base_se_norm": [-0.7536240816116333, -0.6441236734390259], "gt_se_raw": [33.0, 39.0], "gt_se_norm": [-0.6988738775253296, -0.6441236734390259], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005338.pt"}
{"idx": 5340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0WWzgGyAH6Y.mp4", "duration": 219.17729166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment Superman stops the Batmobile.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 193.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment Superman stops the Batmobile.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 193.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [193.0, 198.0], "base_se_norm": [0.7611314058303833, 0.8067564964294434], "gt_se_raw": [190.0, 197.0], "gt_se_norm": [0.7337563037872314, 0.7976315021514893], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005340.pt"}
{"idx": 5342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0WWzgGyAH6Y.mp4", "duration": 219.17729166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man is holding a newspaper that says 'WAYNE TOWER DEVASTATED'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man is holding a newspaper that says 'WAYNE TOWER DEVASTATED'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.5072482228279114, -0.4798731207847595], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [-0.4981231689453125, -0.4798731207847595], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005342.pt"}
{"idx": 5344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yq3Z9msqnOg.mp4", "duration": 162.625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy get up from his seat and run towards the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy get up from his seat and run towards the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [-0.0530361533164978, -0.0038431882858276367], "gt_se_raw": [83.0, 87.0], "gt_se_norm": [0.02075326442718506, 0.06994616985321045], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005344.pt"}
{"idx": 5346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yq3Z9msqnOg.mp4", "duration": 162.625, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the boy is crying at his desk after a flashback.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the boy is crying at his desk after a flashback.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.5695618391036987, -0.47117602825164795], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.5572636127471924, -0.5203689336776733], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005346.pt"}
{"idx": 5348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ER8yEOY0NCc.mp4", "duration": 49.3493, "prompt_text": "Please find the visual event described by the sentence 'At what point is the child rubbing his eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the child rubbing his eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.8784177303314209], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9189451336860657], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005348.pt"}
{"idx": 5350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ER8yEOY0NCc.mp4", "duration": 49.3493, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman uses sign language, gesturing 'finished'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman uses sign language, gesturing 'finished'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [0.5805695056915283, 0.6616243124008179], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.09424042701721191, 0.17529523372650146], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005350.pt"}
{"idx": 5352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ER8yEOY0NCc.mp4", "duration": 49.3493, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the child's hand into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the child's hand into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.5541983842849731, -0.4731435179710388], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [0.7021517753601074, 0.8642615079879761], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005352.pt"}
{"idx": 5354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/foN-gLb27jY.mp4", "duration": 153.754, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man tap the bowl on the boy's head like a drum?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man tap the bowl on the boy's head like a drum?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.4796883463859558, -0.3235948085784912], "gt_se_raw": [41.0, 52.0], "gt_se_norm": [-0.46668052673339844, -0.3235948085784912], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005354.pt"}
{"idx": 5356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/foN-gLb27jY.mp4", "duration": 153.754, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man puts the oven mitt on the boy's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man puts the oven mitt on the boy's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 53.0], "base_se_norm": [-0.6748051643371582, -0.3105870485305786], "gt_se_raw": [110.0, 122.0], "gt_se_norm": [0.4308570623397827, 0.5869506597518921], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005356.pt"}
{"idx": 5358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wEp4ItQzzCU.mp4", "duration": 129.26246666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the puppy barking for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the puppy barking for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [-0.39657658338546753, -0.36563175916671753], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.7214968800544739, -0.6750797033309937], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005358.pt"}
{"idx": 5360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wEp4ItQzzCU.mp4", "duration": 129.26246666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the puppy give a high-five and get praised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the puppy give a high-five and get praised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.45846617221832275, -0.36563175916671753], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [-0.04071146249771118, 0.021178126335144043], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005360.pt"}
{"idx": 5362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wEp4ItQzzCU.mp4", "duration": 129.26246666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the part where the puppy gives a high-five and the person says 'Yay'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the part where the puppy gives a high-five and the person says 'Yay'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.45846617221832275, -0.36563175916671753], "gt_se_raw": [121.0, 124.0], "gt_se_norm": [0.8721598386764526, 0.9185770750045776], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005362.pt"}
{"idx": 5364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZgwNixjNyzI.mp4", "duration": 29.829, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a girl is playing the drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a girl is playing the drum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.3965604305267334, -0.2624627351760864], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.3965604305267334, -0.2624627351760864], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005364.pt"}
{"idx": 5366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZgwNixjNyzI.mp4", "duration": 29.829, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for the scene with the child on the swing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for the scene with the child on the swing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.27392804622650146, 0.47507452964782715], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.27392804622650146, 0.47507452964782715], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005366.pt"}
{"idx": 5368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUeiWQtcPYM.mp4", "duration": 113.48014681348015, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman giving a presentation in a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman giving a presentation in a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 17.0], "base_se_norm": [-0.8590061664581299, -0.7003881335258484], "gt_se_raw": [8.0, 17.0], "gt_se_norm": [-0.8590061664581299, -0.7003881335258484], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005368.pt"}
{"idx": 5370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUeiWQtcPYM.mp4", "duration": 113.48014681348015, "prompt_text": "Please find the visual event described by the sentence 'At what point is a course curriculum for 'Department of Design & Merchandising' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a course curriculum for 'Department of Design & Merchandising' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 47.0], "base_se_norm": [-0.36552780866622925, -0.17166125774383545], "gt_se_raw": [35.0, 47.0], "gt_se_norm": [-0.3831520080566406, -0.17166125774383545], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005370.pt"}
{"idx": 5372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUeiWQtcPYM.mp4", "duration": 113.48014681348015, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman introduces herself by name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman introduces herself by name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9823758006095886, -0.9295030832290649], "gt_se_raw": [105.0, 109.0], "gt_se_norm": [0.8505439758300781, 0.9210408926010132], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005372.pt"}
{"idx": 5374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F5B9jmB_4oM.mp4", "duration": 136.46966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone holding up a lantern?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone holding up a lantern?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 30.0], "base_se_norm": [-0.5749971270561218, -0.5603418350219727], "gt_se_raw": [30.0, 31.0], "gt_se_norm": [-0.5603418350219727, -0.5456866025924683], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005374.pt"}
{"idx": 5376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F5B9jmB_4oM.mp4", "duration": 136.46966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man taking off his hood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man taking off his hood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 74.0], "base_se_norm": [0.04052424430847168, 0.08449006080627441], "gt_se_raw": [73.0, 74.0], "gt_se_norm": [0.06983482837677002, 0.08449006080627441], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005376.pt"}
{"idx": 5378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CU7VAIup27U.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman taking Yorkshire puddings out of the freezer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman taking Yorkshire puddings out of the freezer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.8999999761581421, -0.800000011920929], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.8999999761581421, -0.800000011920929], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005378.pt"}
{"idx": 5380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CU7VAIup27U.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does a little boy try to sneak a piece of food from the kitchen counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a little boy try to sneak a piece of food from the kitchen counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.1666666865348816, -0.06666666269302368], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.1666666865348816, -0.10000002384185791], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005380.pt"}
{"idx": 5382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CU7VAIup27U.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where gravy is being poured over a plate of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where gravy is being poured over a plate of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [0.8666666746139526, 0.9666666984558105], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [0.8666666746139526, 0.9666666984558105], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005382.pt"}
{"idx": 5384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zd2xn5U6e-E.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the green-haired woman holds up a magnifying glass to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the green-haired woman holds up a magnifying glass to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [-0.5868942141532898, -0.5551168918609619], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.5763018131256104, -0.5551168918609619], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005384.pt"}
{"idx": 5386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zd2xn5U6e-E.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the green-haired woman pulls a large watering can from a bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the green-haired woman pulls a large watering can from a bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 70.0], "base_se_norm": [-0.2797130346298218, -0.25852811336517334], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.2797130346298218, -0.25852811336517334], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005386.pt"}
{"idx": 5388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eaIvk1cSyG8.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy walking from the street to the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy walking from the street to the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.5188155174255371, -0.3337445855140686], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.5188155174255371, -0.407772958278656], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005388.pt"}
{"idx": 5390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eaIvk1cSyG8.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the boy says he feels happy with himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the boy says he feels happy with himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [0.4435533285140991, 0.5916100740432739], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.8149290680885315, -0.7409006357192993], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005390.pt"}
{"idx": 5392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eaIvk1cSyG8.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy raise his arms and shout 'rock and roll'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy raise his arms and shout 'rock and roll'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.7766810655593872, 0.924737811088562], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [0.8507094383239746, 0.924737811088562], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005392.pt"}
{"idx": 5394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/17W4QMNwm-c.mp4", "duration": 89.55613333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man with the long hair is kneeling on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man with the long hair is kneeling on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 72.0], "base_se_norm": [0.07195329666137695, 0.607930064201355], "gt_se_raw": [48.0, 71.0], "gt_se_norm": [0.07195329666137695, 0.5855976343154907], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005394.pt"}
{"idx": 5396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/17W4QMNwm-c.mp4", "duration": 89.55613333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the fighters' lightsabers are crossed and locked together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the fighters' lightsabers are crossed and locked together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [-0.1737026572227478, -0.06204080581665039], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.1737026572227478, -0.06204080581665039], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005396.pt"}
{"idx": 5398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z3ZAGBL6UBA.mp4", "duration": 106.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the phrase 'peanut butter jelly with a baseball bat' first get mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the phrase 'peanut butter jelly with a baseball bat' first get mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.679497241973877, -0.6229379177093506], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.641791045665741, -0.5852317810058594], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005398.pt"}
{"idx": 5400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z3ZAGBL6UBA.mp4", "duration": 106.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the song says 'Now break it down and freeze'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the song says 'Now break it down and freeze'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.43440693616867065, -0.3589945435523987], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.17046350240707397, -0.11390417814254761], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005400.pt"}
{"idx": 5402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z3ZAGBL6UBA.mp4", "duration": 106.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the banana starts doing the 'freestyle' dance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the banana starts doing the 'freestyle' dance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9057344794273376, -0.8114689588546753], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.7721916437149048, 0.8287509679794312], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005402.pt"}
{"idx": 5404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CrqIVVd3hp0.mp4", "duration": 154.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the boy in the powered wheelchair watches an advertisement on TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the boy in the powered wheelchair watches an advertisement on TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.7542043924331665, -0.6895213723182678], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.7283312082290649, -0.6895213723182678], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005404.pt"}
{"idx": 5406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CrqIVVd3hp0.mp4", "duration": 154.6, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the boy in the powered wheelchair is happily playing on the accessible swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the boy in the powered wheelchair is happily playing on the accessible swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 147.0], "base_se_norm": [0.8369985818862915, 0.901681661605835], "gt_se_raw": [143.0, 147.0], "gt_se_norm": [0.8499352931976318, 0.901681661605835], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005406.pt"}
{"idx": 5408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MId2A3Mdyn0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the female teacher standing at the podium and giving a lecture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the female teacher standing at the podium and giving a lecture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005408.pt"}
{"idx": 5410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MId2A3Mdyn0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a young man is studying at a computer with a calculator on his desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a young man is studying at a computer with a calculator on his desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.016300976276397705, 0.44275856018066406], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005410.pt"}
{"idx": 5412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MId2A3Mdyn0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where we see a student in a pink sweater taking notes in the foreground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where we see a student in a pink sweater taking notes in the foreground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005412.pt"}
{"idx": 5414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i3kIpCzLzEo.mp4", "duration": 54.053999999999995, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the two adult fish are teaching the baby fish to play hide-and-seek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the two adult fish are teaching the baby fish to play hide-and-seek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 34.0], "base_se_norm": [-0.7039996981620789, 0.2580012083053589], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.37099939584732056, -0.18599921464920044], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005414.pt"}
{"idx": 5416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i3kIpCzLzEo.mp4", "duration": 54.053999999999995, "prompt_text": "Please find the visual event described by the sentence 'When does the baby fish get distracted by the sand while counting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby fish get distracted by the sand while counting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 44.0], "base_se_norm": [0.443001389503479, 0.6280015707015991], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [0.5540015697479248, 0.665001630783081], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005416.pt"}
{"idx": 5418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uEu6r8MkQ0o.mp4", "duration": 152.08541875208542, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the woman is speaking in front of a whiteboard to an audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the woman is speaking in front of a whiteboard to an audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 20.0], "base_se_norm": [-0.8553444147109985, -0.7369899153709412], "gt_se_raw": [11.0, 20.0], "gt_se_norm": [-0.8553444147109985, -0.7369899153709412], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005418.pt"}
{"idx": 5420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uEu6r8MkQ0o.mp4", "duration": 152.08541875208542, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman pushing a trolley through a room full of wooden barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman pushing a trolley through a room full of wooden barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [-0.276722252368927, -0.23727071285247803], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.28987276554107666, -0.23727071285247803], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005420.pt"}
{"idx": 5422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uEu6r8MkQ0o.mp4", "duration": 152.08541875208542, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a collection of lab glassware is shown drying on a black rack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a collection of lab glassware is shown drying on a black rack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 101.0], "base_se_norm": [0.27559900283813477, 0.32820093631744385], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [0.27559900283813477, 0.31505048274993896], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005422.pt"}
{"idx": 5424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tDRBb0apDg.mp4", "duration": 154.02053333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where one of the men is wearing a blonde wig.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where one of the men is wearing a blonde wig.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 55.0], "base_se_norm": [-0.31178009510040283, -0.2858095169067383], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [-0.31178009510040283, -0.2858095169067383], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005424.pt"}
{"idx": 5426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tDRBb0apDg.mp4", "duration": 154.02053333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man hold up the Blu-ray case for the movie 'Up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man hold up the Blu-ray case for the movie 'Up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 82.0], "base_se_norm": [0.012851953506469727, 0.06479310989379883], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.012851953506469727, 0.05180776119232178], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005426.pt"}
{"idx": 5428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VFbYadm_mrw.mp4", "duration": 122.03858333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the cartoon illustration for 'Smell-o-vision'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the cartoon illustration for 'Smell-o-vision'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.3772461414337158, -0.3280813694000244], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.3772461414337158, -0.3280813694000244], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005428.pt"}
{"idx": 5430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VFbYadm_mrw.mp4", "duration": 122.03858333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman in the hat using a tablet to detect the aroma of daisies?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman in the hat using a tablet to detect the aroma of daisies?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [-0.1478104591369629, -0.06586921215057373], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [-0.1478104591369629, -0.06586921215057373], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005430.pt"}
{"idx": 5432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VFbYadm_mrw.mp4", "duration": 122.03858333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a group of people are smelling their laptops around a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a group of people are smelling their laptops around a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 99.0], "base_se_norm": [0.507719874382019, 0.6224377155303955], "gt_se_raw": [92.0, 94.0], "gt_se_norm": [0.507719874382019, 0.5404964685440063], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005432.pt"}
{"idx": 5434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sNcJMejrcnM.mp4", "duration": 104.23746666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the boy is shown interacting with a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the boy is shown interacting with a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 28.0], "base_se_norm": [-0.8848782181739807, -0.4627651572227478], "gt_se_raw": [7.0, 28.0], "gt_se_norm": [-0.8656913042068481, -0.4627651572227478], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005434.pt"}
{"idx": 5436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sNcJMejrcnM.mp4", "duration": 104.23746666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time are other children shown gathering closely around the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are other children shown gathering closely around the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 74.0], "base_se_norm": [0.1704045534133911, 0.4198349714279175], "gt_se_raw": [60.0, 74.0], "gt_se_norm": [0.15121757984161377, 0.4198349714279175], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005436.pt"}
{"idx": 5438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4GfBVEoxStA.mp4", "duration": 156.48966666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the presenter giving her speech on the television screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the presenter giving her speech on the television screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 45.0], "base_se_norm": [-0.9744392037391663, -0.424882173538208], "gt_se_raw": [2.0, 45.0], "gt_se_norm": [-0.9744392037391663, -0.424882173538208], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005438.pt"}
{"idx": 5440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4GfBVEoxStA.mp4", "duration": 156.48966666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video analyze disability representation in 'Finding Nemo'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video analyze disability representation in 'Finding Nemo'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 134.0], "base_se_norm": [0.1885768175125122, 0.7125730514526367], "gt_se_raw": [92.0, 134.0], "gt_se_norm": [0.1757965087890625, 0.7125730514526367], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005440.pt"}
{"idx": 5442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4GfBVEoxStA.mp4", "duration": 156.48966666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the concluding part of the presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the concluding part of the presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 156.0], "base_se_norm": [0.725353479385376, 0.9937418699264526], "gt_se_raw": [134.0, 155.0], "gt_se_norm": [0.7125730514526367, 0.9809614419937134], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005442.pt"}
{"idx": 5444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MdEV-jWMGWw.mp4", "duration": 190.52385719052387, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the pilot pretends there's an emergency, and the woman starts to read the fake checklist.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the pilot pretends there's an emergency, and the woman starts to read the fake checklist.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 103.0], "base_se_norm": [-0.43314188718795776, 0.0812293291091919], "gt_se_raw": [72.0, 92.0], "gt_se_norm": [-0.24418914318084717, -0.03424173593521118], "start_token_span": [77, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005444.pt"}
{"idx": 5446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MdEV-jWMGWw.mp4", "duration": 190.52385719052387, "prompt_text": "Please find the visual event described by the sentence 'Show the woman's reaction after she realizes it was a proposal prank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman's reaction after she realizes it was a proposal prank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 158.0], "base_se_norm": [0.5011242628097534, 0.6585848331451416], "gt_se_raw": [142.0, 158.0], "gt_se_norm": [0.4906269311904907, 0.6585848331451416], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005446.pt"}
{"idx": 5448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pz70gsQVr40.mp4", "duration": 188.6050833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the three owners of the candy shop introduced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three owners of the candy shop introduced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.7454999685287476, -0.6606665849685669], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.7454999685287476, -0.692479133605957], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005448.pt"}
{"idx": 5450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pz70gsQVr40.mp4", "duration": 188.6050833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put a gummy in the shape of teeth into her mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put a gummy in the shape of teeth into her mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 69.0], "base_se_norm": [-0.3001249432563782, -0.26831239461898804], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [-0.3001249432563782, -0.26831239461898804], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005450.pt"}
{"idx": 5452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pz70gsQVr40.mp4", "duration": 188.6050833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where someone is scooping up candies shaped like sunflower seeds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where someone is scooping up candies shaped like sunflower seeds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [-0.5334166288375854, -0.5016040802001953], "gt_se_raw": [138.0, 140.0], "gt_se_norm": [0.4633752107620239, 0.4845834970474243], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005452.pt"}
{"idx": 5454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtP83CWOMwc.mp4", "duration": 198.45833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the wolf blow down the house of straw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the wolf blow down the house of straw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 85.0], "base_se_norm": [-0.18370771408081055, -0.14339697360992432], "gt_se_raw": [82.0, 85.0], "gt_se_norm": [-0.17363005876541138, -0.14339697360992432], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005454.pt"}
{"idx": 5456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtP83CWOMwc.mp4", "duration": 198.45833333333334, "prompt_text": "Please find the visual event described by the sentence 'When do all three pigs sing together inside the brick house for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all three pigs sing together inside the brick house for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 137.0], "base_se_norm": [0.3100987672805786, 0.38064253330230713], "gt_se_raw": [131.0, 137.0], "gt_se_norm": [0.320176362991333, 0.38064253330230713], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005456.pt"}
{"idx": 5458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLeZc7li5HM.mp4", "duration": 96.096, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher guiding the student to fold the cardboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher guiding the student to fold the cardboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.8334998488426208], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9583749771118164], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005458.pt"}
{"idx": 5460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLeZc7li5HM.mp4", "duration": 96.096, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the teacher places the student's hand on her throat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the teacher places the student's hand on her throat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.12587416172027588, -0.042624056339263916], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.14668667316436768, -0.08424907922744751], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005460.pt"}
{"idx": 5462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLeZc7li5HM.mp4", "duration": 96.096, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher start laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher start laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [0.12387609481811523, 0.29037630558013916], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.12387609481811523, 0.16550111770629883], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005462.pt"}
{"idx": 5464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZG4coPHkTmQ.mp4", "duration": 112.57913333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the first bird arrives at the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the first bird arrives at the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 36.0], "base_se_norm": [-0.467041552066803, -0.3604498505592346], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.467041552066803, -0.4315109848976135], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005464.pt"}
{"idx": 5466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZG4coPHkTmQ.mp4", "duration": 112.57913333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the egg starts cracking and the new chick comes out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the egg starts cracking and the new chick comes out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.040674805641174316, 0.11921274662017822], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.3856920003890991, 0.4389878511428833], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005466.pt"}
{"idx": 5468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/glQ2NrnDHWc.mp4", "duration": 180.16, "prompt_text": "Please find the visual event described by the sentence 'When does the blue balloon burst on the single pin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue balloon burst on the single pin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 22.0], "base_se_norm": [-0.8112788796424866, -0.7557726502418518], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.8112788796424866, -0.7890763878822327], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005468.pt"}
{"idx": 5470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/glQ2NrnDHWc.mp4", "duration": 180.16, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the vinegar is being poured into the plastic bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the vinegar is being poured into the plastic bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 64.0], "base_se_norm": [-0.33392542600631714, -0.2895204424858093], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.33392542600631714, -0.3006216883659363], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005470.pt"}
{"idx": 5472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/glQ2NrnDHWc.mp4", "duration": 180.16, "prompt_text": "Please find the visual event described by the sentence 'When is the purple water-filled balloon held over the candle flame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the purple water-filled balloon held over the candle flame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 142.0], "base_se_norm": [0.44316160678863525, 0.5763765573501587], "gt_se_raw": [130.0, 143.0], "gt_se_norm": [0.44316160678863525, 0.5874778032302856], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005472.pt"}
{"idx": 5474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G4Sn91t1V4g.mp4", "duration": 177.09358333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman sleeping in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman sleeping in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.6386091709136963, -0.5934352874755859], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.6386091709136963, -0.5934352874755859], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005474.pt"}
{"idx": 5476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G4Sn91t1V4g.mp4", "duration": 177.09358333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a person is vacuuming the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a person is vacuuming the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 127.0], "base_se_norm": [0.36650919914245605, 0.43426990509033203], "gt_se_raw": [118.0, 127.0], "gt_se_norm": [0.3326287269592285, 0.43426990509033203], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005476.pt"}
{"idx": 5478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rUdA54Xk8cg.mp4", "duration": 106.02258333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does Pinocchio's nose grow for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Pinocchio's nose grow for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.7170413732528687, -0.660449743270874], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.6981775164604187, -0.660449743270874], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005478.pt"}
{"idx": 5480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rUdA54Xk8cg.mp4", "duration": 106.02258333333333, "prompt_text": "Please find the visual event described by the sentence 'When can you see two blue birds sitting in a nest on Pinocchio's nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can you see two blue birds sitting in a nest on Pinocchio's nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.0945325493812561, -0.0002130270004272461], "gt_se_raw": [50.0, 63.0], "gt_se_norm": [-0.05680471658706665, 0.18842601776123047], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005480.pt"}
{"idx": 5482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rUdA54Xk8cg.mp4", "duration": 106.02258333333333, "prompt_text": "Please find the visual event described by the sentence 'When does Jiminy Cricket run down Pinocchio's nose towards his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jiminy Cricket run down Pinocchio's nose towards his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.3774911165237427, -0.30203551054000854], "gt_se_raw": [74.0, 76.0], "gt_se_norm": [0.3959289789199829, 0.43365681171417236], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005482.pt"}
{"idx": 5484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U9Ots7t3_4g.mp4", "duration": 199.63276666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the teacher guides the child's hand to touch the year 2009.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the teacher guides the child's hand to touch the year 2009.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 46.0], "base_se_norm": [-0.6593745946884155, -0.5391538143157959], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.6293193697929382, -0.5892457962036133], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005484.pt"}
{"idx": 5486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U9Ots7t3_4g.mp4", "duration": 199.63276666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher hold the child's head with both hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher hold the child's head with both hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 120.0], "base_se_norm": [0.1621338129043579, 0.20220744609832764], "gt_se_raw": [116.0, 120.0], "gt_se_norm": [0.1621338129043579, 0.20220744609832764], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005486.pt"}
{"idx": 5488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z1fbwPHv-wA.mp4", "duration": 68.9689, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man adjusts the boy's grip on the cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man adjusts the boy's grip on the cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [0.5369246006011963, 0.6819174289703369], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.9130042791366577, -0.7680113911628723], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005488.pt"}
{"idx": 5490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z1fbwPHv-wA.mp4", "duration": 68.9689, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man helps the boy pick up the cane from the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man helps the boy pick up the cane from the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [-0.10104411840438843, 0.04394876956939697], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.07204556465148926, 0.07294738292694092], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005490.pt"}
{"idx": 5492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z1fbwPHv-wA.mp4", "duration": 68.9689, "prompt_text": "Please find the visual event described by the sentence 'When are the man and the boy walking together across the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and the boy walking together across the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 31.0], "base_se_norm": [-0.4200284481048584, -0.10104411840438843], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.7109160423278809, 0.8269103765487671], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005492.pt"}
{"idx": 5494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f-B-640jgWY.mp4", "duration": 166.53303333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman speaking in front of a quilt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman speaking in front of a quilt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 102.0], "base_se_norm": [0.15292441844940186, 0.2249821424484253], "gt_se_raw": [96.0, 103.0], "gt_se_norm": [0.15292441844940186, 0.2369917631149292], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005494.pt"}
{"idx": 5496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f-B-640jgWY.mp4", "duration": 166.53303333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man appear with a dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man appear with a dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 144.0], "base_se_norm": [0.6453192234039307, 0.729386568069458], "gt_se_raw": [137.0, 145.0], "gt_se_norm": [0.6453192234039307, 0.7413961887359619], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005496.pt"}
{"idx": 5498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i07EkgDQvy8.mp4", "duration": 132.52, "prompt_text": "Please find the visual event described by the sentence 'Show me when the two men are hanging laundry on the clothesline and talking to each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the two men are hanging laundry on the clothesline and talking to each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 83.0], "base_se_norm": [-0.366133451461792, 0.25264108180999756], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [-0.1246604323387146, -0.019016027450561523], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005498.pt"}
{"idx": 5500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i07EkgDQvy8.mp4", "duration": 132.52, "prompt_text": "Please find the visual event described by the sentence 'When does one of the men carry a wooden bucket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does one of the men carry a wooden bucket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 44.0], "base_se_norm": [-0.3963175415992737, -0.3359493017196655], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.3963175415992737, -0.35104137659072876], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005500.pt"}
{"idx": 5502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i07EkgDQvy8.mp4", "duration": 132.52, "prompt_text": "Please find the visual event described by the sentence 'At what point do the men discuss 'sex equality'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the men discuss 'sex equality'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 118.0], "base_se_norm": [0.599758505821228, 0.7808631658554077], "gt_se_raw": [114.0, 118.0], "gt_se_norm": [0.7204949855804443, 0.7808631658554077], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005502.pt"}
{"idx": 5504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6oyQO-H9YxM.mp4", "duration": 107.732625, "prompt_text": "Please find the visual event described by the sentence 'When are the Smart Seats first introduced and shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the Smart Seats first introduced and shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 21.0], "base_se_norm": [-0.7029683589935303, -0.6101459264755249], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.6844038963317871, -0.6101459264755249], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005504.pt"}
{"idx": 5506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6oyQO-H9YxM.mp4", "duration": 107.732625, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a man on crutches almost gets hit by the moving seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a man on crutches almost gets hit by the moving seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 67.0], "base_se_norm": [0.15099763870239258, 0.24382007122039795], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [0.13243317604064941, 0.16956210136413574], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005506.pt"}
{"idx": 5508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGm_DqYtm_Q.mp4", "duration": 11.311, "prompt_text": "Please find the visual event described by the sentence 'When is the person brushing their hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person brushing their hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.9450093507766724], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.9450093507766724], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005508.pt"}
{"idx": 5510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGm_DqYtm_Q.mp4", "duration": 11.311, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a person in a purple shirt is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a person in a purple shirt is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.9450093507766724], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.9450093507766724], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005510.pt"}
{"idx": 5512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGm_DqYtm_Q.mp4", "duration": 11.311, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the person's arm moves up and down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the person's arm moves up and down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.7681902647018433], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.9450093507766724], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005512.pt"}
{"idx": 5514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-6rP8FO_Vwc.mp4", "duration": 364.5308333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the Big Macs and Filet-O-Fish sandwiches revealed for lunch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the Big Macs and Filet-O-Fish sandwiches revealed for lunch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 88.0], "base_se_norm": [-0.5939986705780029, -0.5171875953674316], "gt_se_raw": [77.0, 80.0], "gt_se_norm": [-0.5775391459465027, -0.5610796213150024], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005514.pt"}
{"idx": 5516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-6rP8FO_Vwc.mp4", "duration": 364.5308333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman scream in reaction to the amount of ketchup being poured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman scream in reaction to the amount of ketchup being poured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 116.0], "base_se_norm": [-0.37453848123550415, -0.36356550455093384], "gt_se_raw": [113.0, 115.0], "gt_se_norm": [-0.3800249695777893, -0.369051992893219], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005516.pt"}
{"idx": 5518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/stoYe8VhIcY.mp4", "duration": 49.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the speaker talking to the camera at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the speaker talking to the camera at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.7981157302856445], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.7981157302856445], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005518.pt"}
{"idx": 5520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/stoYe8VhIcY.mp4", "duration": 49.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the speaker shows a plant that isn't doing well.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the speaker shows a plant that isn't doing well.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 38.0], "base_se_norm": [-0.07133245468139648, 0.5343203544616699], "gt_se_raw": [24.0, 38.0], "gt_se_norm": [-0.03095555305480957, 0.5343203544616699], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005520.pt"}
{"idx": 5522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/stoYe8VhIcY.mp4", "duration": 49.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a digital clock is next to the plant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a digital clock is next to the plant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.636608362197876, -0.47510093450546265], "gt_se_raw": [10.0, 23.0], "gt_se_norm": [-0.5962314605712891, -0.07133245468139648], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005522.pt"}
{"idx": 5524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6TEwKxBs78k.mp4", "duration": 160.3602, "prompt_text": "Please find the visual event described by the sentence 'When is the character mining stone blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the character mining stone blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 37.0], "base_se_norm": [-0.7131457924842834, -0.5385388731956482], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.5884265899658203, -0.55101078748703], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005524.pt"}
{"idx": 5526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6TEwKxBs78k.mp4", "duration": 160.3602, "prompt_text": "Please find the visual event described by the sentence 'When is the character in the brown outfit looking at the cobweb?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the character in the brown outfit looking at the cobweb?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 90.0], "base_se_norm": [0.06011343002319336, 0.12247300148010254], "gt_se_raw": [85.0, 89.0], "gt_se_norm": [0.06011343002319336, 0.1100010871887207], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005526.pt"}
{"idx": 5528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M1GO1X09Gec.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a nurse taking another person's blood pressure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a nurse taking another person's blood pressure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005528.pt"}
{"idx": 5530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M1GO1X09Gec.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where students are attending a class with a teacher at the front.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where students are attending a class with a teacher at the front.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.016300976276397705, 0.44275856018066406], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005530.pt"}
{"idx": 5532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M1GO1X09Gec.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the text 'MEDICAL ASSISTING' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the text 'MEDICAL ASSISTING' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9344200491905212, -0.5409404635429382], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9344200491905212, -0.5409404635429382], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005532.pt"}
{"idx": 5534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fCFkOtlE9zg.mp4", "duration": 54.304249999999996, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the boat sinks completely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the boat sinks completely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.3002389073371887, -0.1897503137588501], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.33706843852996826, -0.1897503137588501], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005534.pt"}
{"idx": 5536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fCFkOtlE9zg.mp4", "duration": 54.304249999999996, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the heart character attempts to spear the brain character.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the heart character attempts to spear the brain character.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [0.06805634498596191, 0.1785449981689453], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.06805634498596191, 0.1785449981689453], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005536.pt"}
{"idx": 5538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-AFUpW3oNg.mp4", "duration": 47.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'When are the wooden pieces still unpolished and stacked on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the wooden pieces still unpolished and stacked on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, -0.2772365212440491], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, -0.48981398344039917], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005538.pt"}
{"idx": 5540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-AFUpW3oNg.mp4", "duration": 47.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the finished and polished wooden piece.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the finished and polished wooden piece.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 46.0], "base_se_norm": [-0.10717451572418213, 0.9557129144668579], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [-0.10717451572418213, 0.1479184627532959], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005540.pt"}
{"idx": 5542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-AFUpW3oNg.mp4", "duration": 47.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see a variety of wooden components scattered on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a variety of wooden components scattered on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, -0.2772365212440491], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.8299379944801331, -0.6173604726791382], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005542.pt"}
{"idx": 5544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_La7IMssNOA.mp4", "duration": 73.70704037370704, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person in a white hazmat suit performing a test on someone in a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person in a white hazmat suit performing a test on someone in a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 48.0], "base_se_norm": [0.058243513107299805, 0.3024536371231079], "gt_se_raw": [44.0, 49.0], "gt_se_norm": [0.19391584396362305, 0.32958805561065674], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005544.pt"}
{"idx": 5546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_La7IMssNOA.mp4", "duration": 73.70704037370704, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a campus directory sign is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a campus directory sign is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.3216387629508972, -0.24023538827896118], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.3216387629508972, -0.24023538827896118], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005546.pt"}
{"idx": 5548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihlN5nf1qew.mp4", "duration": 94.96987499999999, "prompt_text": "Please find the visual event described by the sentence 'When does Jim successfully prank Michael with the 'updog' joke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jim successfully prank Michael with the 'updog' joke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.620932400226593, -0.45245790481567383], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.7262289524078369, -0.620932400226593], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005548.pt"}
{"idx": 5550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihlN5nf1qew.mp4", "duration": 94.96987499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Michael fails to prank Ryan in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Michael fails to prank Ryan in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.2629241347312927, -0.09444963932037354], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [-0.2629241347312927, -0.09444963932037354], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005550.pt"}
{"idx": 5552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihlN5nf1qew.mp4", "duration": 94.96987499999999, "prompt_text": "Please find the visual event described by the sentence 'When does Jim give a subtle look to the camera after Michael's repeated failed attempts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jim give a subtle look to the camera after Michael's repeated failed attempts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 73.0], "base_se_norm": [0.28461790084838867, 0.5373296737670898], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [0.49521100521087646, 0.5373296737670898], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005552.pt"}
{"idx": 5554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dL1kDLQ8-kg.mp4", "duration": 41.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a single snowball rolls and then stops.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a single snowball rolls and then stops.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [-0.22580641508102417, -0.12903225421905518], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.22580641508102417, -0.12903225421905518], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005554.pt"}
{"idx": 5556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dL1kDLQ8-kg.mp4", "duration": 41.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'When does Olaf pat the head of the little snowman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Olaf pat the head of the little snowman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 30.0], "base_se_norm": [0.3548387289047241, 0.4516129493713379], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.3548387289047241, 0.4516129493713379], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005556.pt"}
{"idx": 5558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T4aWsD3jW1w.mp4", "duration": 68.70196666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy reacting emotionally to the violin music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy reacting emotionally to the violin music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 16.0], "base_se_norm": [-0.9708887338638306, -0.5342199802398682], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9708887338638306, -0.5633312463760376], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005558.pt"}
{"idx": 5560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T4aWsD3jW1w.mp4", "duration": 68.70196666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the birthday cake is brought to the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the birthday cake is brought to the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [0.2808949947357178, 0.3682286739349365], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [0.25178372859954834, 0.3682286739349365], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005560.pt"}
{"idx": 5562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T4aWsD3jW1w.mp4", "duration": 68.70196666666666, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a blue shirt assist the boy in holding the ice cream cone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a blue shirt assist the boy in holding the ice cream cone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 67.0], "base_se_norm": [0.7466748952865601, 0.9504536390304565], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.8048974275588989, 0.9504536390304565], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005562.pt"}
{"idx": 5564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMb4fgiBimU.mp4", "duration": 81.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman guides the man to feel the texture of a headstone with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman guides the man to feel the texture of a headstone with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 10.0], "base_se_norm": [-0.9506375789642334, -0.7531880140304565], "gt_se_raw": [2.0, 9.0], "gt_se_norm": [-0.9506375789642334, -0.7778692245483398], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005564.pt"}
{"idx": 5566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMb4fgiBimU.mp4", "duration": 81.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time do they examine the headstone of 'JUANITA E', the wife of a soldier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do they examine the headstone of 'JUANITA E', the wife of a soldier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 58.0], "base_se_norm": [-0.03743314743041992, 0.43150973320007324], "gt_se_raw": [41.0, 54.0], "gt_se_norm": [0.01192927360534668, 0.33278489112854004], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005566.pt"}
{"idx": 5568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cWX9iR_PzIQ.mp4", "duration": 193.52686019352686, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the group of friends walking into the mall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the group of friends walking into the mall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 26.0], "base_se_norm": [-0.8863207101821899, -0.7313034534454346], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.8759862184524536, -0.855317234992981], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005568.pt"}
{"idx": 5570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cWX9iR_PzIQ.mp4", "duration": 193.52686019352686, "prompt_text": "Please find the visual event described by the sentence 'When do the kids enter the ice cream parlor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the kids enter the ice cream parlor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [-0.46260690689086914, -0.43160343170166016], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.46260690689086914, -0.4419379234313965], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005570.pt"}
{"idx": 5572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cWX9iR_PzIQ.mp4", "duration": 193.52686019352686, "prompt_text": "Please find the visual event described by the sentence 'When do all the lights in the mall suddenly go out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the lights in the mall suddenly go out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 142.0], "base_se_norm": [0.3951551914215088, 0.4674966335296631], "gt_se_raw": [135.0, 137.0], "gt_se_norm": [0.3951551914215088, 0.41582417488098145], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005572.pt"}
{"idx": 5574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6R-H-pdufJg.mp4", "duration": 91.05772439105773, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a person is speaking in front of a bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a person is speaking in front of a bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 41.0], "base_se_norm": [-0.38500547409057617, -0.09947234392166138], "gt_se_raw": [27.0, 42.0], "gt_se_norm": [-0.4069696068763733, -0.07750821113586426], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005574.pt"}
{"idx": 5576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6R-H-pdufJg.mp4", "duration": 91.05772439105773, "prompt_text": "Please find the visual event described by the sentence 'At what point is the construction of a large industrial facility shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the construction of a large industrial facility shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [0.4276658296585083, 0.49355804920196533], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.4276658296585083, 0.49355804920196533], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005576.pt"}
{"idx": 5578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bp1kg2jFvJM.mp4", "duration": 162.295, "prompt_text": "Please find the visual event described by the sentence 'When can we see a truck with 'LUCKY LAGER' written on its side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a truck with 'LUCKY LAGER' written on its side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.987676739692688, -0.9507070183753967], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.987676739692688, -0.9507070183753967], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005578.pt"}
{"idx": 5580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bp1kg2jFvJM.mp4", "duration": 162.295, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman tell another woman that she is blind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman tell another woman that she is blind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 101.0], "base_se_norm": [0.15838444232940674, 0.24464714527130127], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.18303096294403076, 0.20767736434936523], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005580.pt"}
{"idx": 5582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bp1kg2jFvJM.mp4", "duration": 162.295, "prompt_text": "Please find the visual event described by the sentence 'When does a man in a light-colored suit help the woman up after she falls?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man in a light-colored suit help the woman up after she falls?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 140.0], "base_se_norm": [0.6389907598495483, 0.7252534627914429], "gt_se_raw": [136.0, 142.0], "gt_se_norm": [0.6759604215621948, 0.7498998641967773], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005582.pt"}
{"idx": 5584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ujncmH8YnNc.mp4", "duration": 188.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the rabbit police officer appears and looks around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the rabbit police officer appears and looks around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 28.0], "base_se_norm": [-0.8835722208023071, -0.7036383748054504], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.8835722208023071, -0.8200661540031433], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005584.pt"}
{"idx": 5586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ujncmH8YnNc.mp4", "duration": 188.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the elephant shopkeeper points to the 'refuse service' sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the elephant shopkeeper points to the 'refuse service' sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 68.0], "base_se_norm": [-0.34377068281173706, -0.28026461601257324], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [-0.3331863284111023, -0.28026461601257324], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005586.pt"}
{"idx": 5588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AkKEtjlk6Qs.mp4", "duration": 117.95128461795129, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in a lab coat and black gloves using a pipette?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in a lab coat and black gloves using a pipette?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [-0.593052327632904, -0.5082715749740601], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.593052327632904, -0.5082715749740601], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005588.pt"}
{"idx": 5590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AkKEtjlk6Qs.mp4", "duration": 117.95128461795129, "prompt_text": "Please find the visual event described by the sentence 'Show me the man pointing at the graph on the projection screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man pointing at the graph on the projection screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 69.0], "base_se_norm": [0.13606226444244385, 0.16997456550598145], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [0.11910605430603027, 0.16997456550598145], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005590.pt"}
{"idx": 5592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AkKEtjlk6Qs.mp4", "duration": 117.95128461795129, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the robotic arm picks up the small glass vial.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the robotic arm picks up the small glass vial.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 37.0], "base_se_norm": [-0.40653467178344727, -0.37262237071990967], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.40653467178344727, -0.37262237071990967], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005592.pt"}
{"idx": 5594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNV63n3td1A.mp4", "duration": 22.022000000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy trails the brick wall with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy trails the brick wall with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 21.0], "base_se_norm": [-0.09181725978851318, 0.9071837663650513], "gt_se_raw": [8.0, 21.0], "gt_se_norm": [-0.2734537720680237, 0.9071837663650513], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005594.pt"}
{"idx": 5596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNV63n3td1A.mp4", "duration": 22.022000000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point is the red fire alarm visible on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the red fire alarm visible on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.18263554573059082, 0.45309245586395264], "gt_se_raw": [6.0, 21.0], "gt_se_norm": [-0.45509034395217896, 0.9071837663650513], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005596.pt"}
{"idx": 5598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WdKbSQWyqDM.mp4", "duration": 23.056366666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the baby put its hand on its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby put its hand on its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.045816659927368164, 0.21441519260406494], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [0.04092729091644287, 0.21441519260406494], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005598.pt"}
{"idx": 5600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WdKbSQWyqDM.mp4", "duration": 23.056366666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the baby putting its fingers in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the baby putting its fingers in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.21441519260406494], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, 0.04092729091644287], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005600.pt"}
{"idx": 5602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WdKbSQWyqDM.mp4", "duration": 23.056366666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the adult points at the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the adult points at the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.1325606107711792], "gt_se_raw": [16.0, 17.0], "gt_se_norm": [0.38790297508239746, 0.4746469259262085], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005602.pt"}
{"idx": 5604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yb0vU-31FOY.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'When does the pink spiky ball roll into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the pink spiky ball roll into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8234696984291077, -0.7432286739349365], "gt_se_raw": [6.0, 13.0], "gt_se_norm": [-0.9037107229232788, -0.7913732528686523], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005604.pt"}
{"idx": 5606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yb0vU-31FOY.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman start laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman start laughing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.8395178914070129], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.30992698669433594, -0.2296859622001648], "start_token_span": [69, 72], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005606.pt"}
{"idx": 5608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xeOC6bipMqU.mp4", "duration": 94.88645833333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman kissing the man on his cheek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman kissing the man on his cheek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.36766529083251953, -0.30443185567855835], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.36766529083251953, -0.30443185567855835], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005608.pt"}
{"idx": 5610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xeOC6bipMqU.mp4", "duration": 94.88645833333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man walks out of the frame to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man walks out of the frame to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 52.0], "base_se_norm": [-0.13580924272537231, 0.0960468053817749], "gt_se_raw": [41.0, 52.0], "gt_se_norm": [-0.13580924272537231, 0.0960468053817749], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005610.pt"}
{"idx": 5612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xeOC6bipMqU.mp4", "duration": 94.88645833333332, "prompt_text": "Please find the visual event described by the sentence 'Identify the timeframe where only the shadows of the two individuals are present on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the timeframe where only the shadows of the two individuals are present on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 72.0], "base_se_norm": [0.0960468053817749, 0.5176032781600952], "gt_se_raw": [52.0, 71.0], "gt_se_norm": [0.0960468053817749, 0.4965254068374634], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005612.pt"}
{"idx": 5614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ldDAt-vPCY0.mp4", "duration": 195.61208333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of adding flour, paprika and garlic powder to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of adding flour, paprika and garlic powder to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 24.0], "base_se_norm": [-0.8670839071273804, -0.754616379737854], "gt_se_raw": [13.0, 20.0], "gt_se_norm": [-0.8670839071273804, -0.7955136299133301], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005614.pt"}
{"idx": 5616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ldDAt-vPCY0.mp4", "duration": 195.61208333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the liquid egg mixture is poured into the pie crust.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the liquid egg mixture is poured into the pie crust.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 160.0], "base_se_norm": [0.5745449066162109, 0.6358908414840698], "gt_se_raw": [155.0, 159.0], "gt_se_norm": [0.5847692489624023, 0.6256664991378784], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005616.pt"}
{"idx": 5618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1GXQLnTalcg.mp4", "duration": 109.943, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man applying toothpaste to the toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man applying toothpaste to the toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.5452188849449158, -0.47245389223098755], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [-0.5634101629257202, -0.490645170211792], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005618.pt"}
{"idx": 5620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1GXQLnTalcg.mp4", "duration": 109.943, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show the man brushing his tongue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show the man brushing his tongue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 86.0], "base_se_norm": [0.27338707447052, 0.5644470453262329], "gt_se_raw": [79.0, 84.0], "gt_se_norm": [0.43710827827453613, 0.528064489364624], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005620.pt"}
{"idx": 5622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1GXQLnTalcg.mp4", "duration": 109.943, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the display of 'you will need' items?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the display of 'you will need' items?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.7271313667297363, -0.5997925996780396], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.7271313667297363, -0.5997925996780396], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005622.pt"}
{"idx": 5624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4yG8caPPY1Y.mp4", "duration": 139.12, "prompt_text": "Please find the visual event described by the sentence 'When can we see a house built into a giant stone hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a house built into a giant stone hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.7412306070327759, -0.7124783992767334], "gt_se_raw": [18.0, 19.0], "gt_se_norm": [-0.7412306070327759, -0.7268545031547546], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005624.pt"}
{"idx": 5626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4yG8caPPY1Y.mp4", "duration": 139.12, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Connie, wearing a dark blue jumpsuit and with her hair braided, defeats a holographic opponent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Connie, wearing a dark blue jumpsuit and with her hair braided, defeats a holographic opponent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 113.0], "base_se_norm": [0.4807361364364624, 0.6244969367980957], "gt_se_raw": [112.0, 114.0], "gt_se_norm": [0.6101207733154297, 0.6388729810714722], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005626.pt"}
{"idx": 5628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ExeH85oCNIs.mp4", "duration": 95.56213333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman raise her hand to interrupt the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman raise her hand to interrupt the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.43492263555526733, -0.288421094417572], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.7488545179367065, -0.7069969177246094], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005628.pt"}
{"idx": 5630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ExeH85oCNIs.mp4", "duration": 95.56213333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman suggest getting sheets for a dorm room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman suggest getting sheets for a dorm room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 25.0], "base_se_norm": [-0.7488545179367065, -0.4767801761627197], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [-0.22563469409942627, -0.1628483533859253], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005630.pt"}
{"idx": 5632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ExeH85oCNIs.mp4", "duration": 95.56213333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man explains the meaning of the letters 'C' and 'M' for the system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man explains the meaning of the letters 'C' and 'M' for the system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [0.360371470451355, 0.5068730115890503], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [0.4440866708755493, 0.5487306118011475], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005632.pt"}
{"idx": 5634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1UdI_eoDPKQ.mp4", "duration": 46.88016666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the dogs slide down the slide one after another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the dogs slide down the slide one after another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.44539445638656616, -0.31740856170654297], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.44539445638656616, -0.31740856170654297], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005634.pt"}
{"idx": 5636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1UdI_eoDPKQ.mp4", "duration": 46.88016666666666, "prompt_text": "Please find the visual event described by the sentence 'When does Marshall the fire pup rescue the boy on the cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Marshall the fire pup rescue the boy on the cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [0.10921108722686768, 0.1945350170135498], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [0.023887157440185547, 0.1945350170135498], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005636.pt"}
{"idx": 5638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p6qbJMBKato.mp4", "duration": 106.04, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher say 'Hello my awesome students'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher say 'Hello my awesome students'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9811391830444336, -0.9245567917823792], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9811391830444336, -0.9245567917823792], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005638.pt"}
{"idx": 5640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p6qbJMBKato.mp4", "duration": 106.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment explaining the rule 'Raise your hands if you want to speak'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment explaining the rule 'Raise your hands if you want to speak'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 39.0], "base_se_norm": [-0.37759333848953247, -0.2644284963607788], "gt_se_raw": [33.0, 39.0], "gt_se_norm": [-0.37759333848953247, -0.2644284963607788], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005640.pt"}
{"idx": 5642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p6qbJMBKato.mp4", "duration": 106.04, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the rule about having fun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the rule about having fun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 91.0], "base_se_norm": [0.49000370502471924, 0.7163333892822266], "gt_se_raw": [79.0, 90.0], "gt_se_norm": [0.49000370502471924, 0.6974726915359497], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005642.pt"}
{"idx": 5644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9yciDzo8fGo.mp4", "duration": 140.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a photo of Eddie Redmayne is on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a photo of Eddie Redmayne is on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 45.0], "base_se_norm": [-0.8013714551925659, -0.3615511655807495], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.8013714551925659, -0.7729959487915039], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005644.pt"}
{"idx": 5646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9yciDzo8fGo.mp4", "duration": 140.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when a red X is placed over a drawing of a brick wall that says 'Disability'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when a red X is placed over a drawing of a brick wall that says 'Disability'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [0.06408143043518066, 0.1066446304321289], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.06408143043518066, 0.1066446304321289], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005646.pt"}
{"idx": 5648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kg-EEBIe7Lk.mp4", "duration": 48.506791666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the hand in the red plaid shirt drop the magazine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the hand in the red plaid shirt drop the magazine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.34029853343963623, -0.21660453081130981], "gt_se_raw": [18.0, 19.0], "gt_se_norm": [-0.25783586502075195, -0.21660453081130981], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005648.pt"}
{"idx": 5650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kg-EEBIe7Lk.mp4", "duration": 48.506791666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point do the words 'SUBSCRIBE TO DAD' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the words 'SUBSCRIBE TO DAD' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 48.0], "base_se_norm": [0.19570887088775635, 0.9791043996810913], "gt_se_raw": [29.0, 34.0], "gt_se_norm": [0.19570887088775635, 0.4018656015396118], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005650.pt"}
{"idx": 5652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kg-EEBIe7Lk.mp4", "duration": 48.506791666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we hear the repetitive 'wow' sound effect?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we hear the repetitive 'wow' sound effect?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 48.0], "base_se_norm": [0.19570887088775635, 0.9791043996810913], "gt_se_raw": [17.0, 24.0], "gt_se_norm": [-0.2990671992301941, -0.010447800159454346], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005652.pt"}
{"idx": 5654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j3cEnP8VCvg.mp4", "duration": 19.875, "prompt_text": "Please find the visual event described by the sentence 'At what point does the blue robot with a screen for a head appear on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the blue robot with a screen for a head appear on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.597484290599823, -0.09433960914611816], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.698113203048706, -0.194968581199646], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005654.pt"}
{"idx": 5656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j3cEnP8VCvg.mp4", "duration": 19.875, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy is playing with the box before the robot appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy is playing with the box before the robot appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.597484290599823], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.698113203048706], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005656.pt"}
{"idx": 5658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W-xuEJVEraw.mp4", "duration": 83.3833, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a white suit giving a speech on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a white suit giving a speech on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.16050338745117188, -0.0885465145111084], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.16050338745117188, -0.0885465145111084], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005658.pt"}
{"idx": 5660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W-xuEJVEraw.mp4", "duration": 83.3833, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see several women in red skirts playing drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see several women in red skirts playing drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 77.0], "base_se_norm": [0.3911658525466919, 0.8468925952911377], "gt_se_raw": [65.0, 76.0], "gt_se_norm": [0.5590652227401733, 0.8229069709777832], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005660.pt"}
{"idx": 5662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W-xuEJVEraw.mp4", "duration": 83.3833, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the title 'ROLL LIKE A SCIENTIST' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the title 'ROLL LIKE A SCIENTIST' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.8800718784332275, -0.7841294407844543], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.8800718784332275, -0.7841294407844543], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005662.pt"}
{"idx": 5664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sJzSUK6-O2Q.mp4", "duration": 138.37156666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the person in the green hoodie assembles the yellow paper components.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the person in the green hoodie assembles the yellow paper components.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 98.0], "base_se_norm": [0.17076075077056885, 0.4164760112762451], "gt_se_raw": [12.0, 21.0], "gt_se_norm": [-0.8265539407730103, -0.6964694261550903], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005664.pt"}
{"idx": 5666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sJzSUK6-O2Q.mp4", "duration": 138.37156666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the instructor guides the student's hands to explore the surface of the fully assembled ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the instructor guides the student's hands to explore the surface of the fully assembled ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 44.0], "base_se_norm": [-0.4941157102584839, -0.36403119564056396], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [-0.29176199436187744, -0.2194928526878357], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005666.pt"}
{"idx": 5668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P4vcI-ptEBI.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone spreading frosting on a bone-shaped cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone spreading frosting on a bone-shaped cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.9065666794776917, -0.8338963389396667], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.885803759098053, -0.8338963389396667], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005668.pt"}
{"idx": 5670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P4vcI-ptEBI.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'When are the mini burgers being assembled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the mini burgers being assembled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.6989371180534363, -0.6055037975311279], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.6677926778793335, -0.6158852577209473], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005670.pt"}
{"idx": 5672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P4vcI-ptEBI.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'When is the man petting the German Shepherd while it's lying on its back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man petting the German Shepherd while it's lying on its back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [-0.7300815582275391, -0.6989371180534363], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.7300815582275391, -0.6989371180534363], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005672.pt"}
{"idx": 5674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3T0-olhTvtM.mp4", "duration": 178.8787, "prompt_text": "Please find the visual event described by the sentence 'When does the character create a whirlwind and fly into the air for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character create a whirlwind and fly into the air for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 32.0], "base_se_norm": [-0.7875655293464661, -0.6422156095504761], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.7652040123939514, -0.70930016040802], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005674.pt"}
{"idx": 5676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3T0-olhTvtM.mp4", "duration": 178.8787, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the bird-like character interrupts the woman by shouting 'However!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the bird-like character interrupts the woman by shouting 'However!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 102.0], "base_se_norm": [0.11807620525360107, 0.14043772220611572], "gt_se_raw": [100.0, 101.0], "gt_se_norm": [0.11807620525360107, 0.1292569637298584], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005676.pt"}
{"idx": 5678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GBYDe-ERdiQ.mp4", "duration": 234.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the person is holding up the zebra mildliners.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the person is holding up the zebra mildliners.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.5643507838249207, -0.5301822423934937], "gt_se_raw": [51.0, 55.0], "gt_se_norm": [-0.5643507838249207, -0.5301822423934937], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005678.pt"}
{"idx": 5680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GBYDe-ERdiQ.mp4", "duration": 234.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a black muji eraser is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a black muji eraser is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [-0.3593394160270691, -0.31662869453430176], "gt_se_raw": [76.0, 79.0], "gt_se_norm": [-0.35079723596572876, -0.3251708149909973], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005680.pt"}
{"idx": 5682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GBYDe-ERdiQ.mp4", "duration": 234.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the Apple pencil is being presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 188.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the Apple pencil is being presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 188.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [188.0, 193.0], "base_se_norm": [0.6059225797653198, 0.6486332416534424], "gt_se_raw": [189.0, 192.0], "gt_se_norm": [0.6144647598266602, 0.6400911808013916], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005682.pt"}
{"idx": 5684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VemMC_GkmSk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person in a yellow gown is preparing an IV bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person in a yellow gown is preparing an IV bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5414421558380127, -0.08288431167602539], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005684.pt"}
{"idx": 5686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VemMC_GkmSk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman is retrieving an item from a high shelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman is retrieving an item from a high shelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3756735324859619, 0.8342313766479492], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005686.pt"}
{"idx": 5688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G_QWtUFFAFQ.mp4", "duration": 283.256, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker explain what the device on his hands is called?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker explain what the device on his hands is called?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.908210277557373, -0.8022990226745605], "gt_se_raw": [13.0, 24.0], "gt_se_norm": [-0.908210277557373, -0.8305419683456421], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005688.pt"}
{"idx": 5690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G_QWtUFFAFQ.mp4", "duration": 283.256, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker state that no modifications were made to the iPhone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker state that no modifications were made to the iPhone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 122.0], "base_se_norm": [-0.16683143377304077, -0.13858842849731445], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [-0.5198689699172974, -0.48456519842147827], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005690.pt"}
{"idx": 5692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G_QWtUFFAFQ.mp4", "duration": 283.256, "prompt_text": "Please find the visual event described by the sentence 'When does an incoming phone call interrupt the demonstration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does an incoming phone call interrupt the demonstration?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 126.0], "base_se_norm": [-0.13858842849731445, -0.11034542322158813], "gt_se_raw": [249.0, 273.0], "gt_se_norm": [0.7581268548965454, 0.9275848865509033], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005692.pt"}
{"idx": 5694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OMTsvsnO3Mk.mp4", "duration": 151.88, "prompt_text": "Please find the visual event described by the sentence 'Show the time range for the text 'Let's listen to young people's voices'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range for the text 'Let's listen to young people's voices'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 123.0], "base_se_norm": [0.5801948308944702, 0.6196997165679932], "gt_se_raw": [120.0, 123.0], "gt_se_norm": [0.5801948308944702, 0.6196997165679932], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005694.pt"}
{"idx": 5696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OMTsvsnO3Mk.mp4", "duration": 151.88, "prompt_text": "Please find the visual event described by the sentence 'When does the man wearing a Star Wars shirt say, 'You have a history of victory'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man wearing a Star Wars shirt say, 'You have a history of victory'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [-0.3415854573249817, -0.3020806312561035], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.683961033821106, -0.6576244831085205], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005696.pt"}
{"idx": 5698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JtzsCx0P3tI.mp4", "duration": 35.101733333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the woman explain that the limbic system is primarily non-verbal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman explain that the limbic system is primarily non-verbal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.25929588079452515, 0.5953627824783325], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.886045515537262, -0.7151137590408325], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005698.pt"}
{"idx": 5700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JtzsCx0P3tI.mp4", "duration": 35.101733333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman use her finger to indicate a specific part of the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman use her finger to indicate a specific part of the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.25929588079452515, 0.5953627824783325], "gt_se_raw": [13.0, 29.0], "gt_se_norm": [-0.25929588079452515, 0.6523400545120239], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005700.pt"}
{"idx": 5702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JtzsCx0P3tI.mp4", "duration": 35.101733333333335, "prompt_text": "Please find the visual event described by the sentence 'When does she use her thumb in a gesture to illustrate 'building up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does she use her thumb in a gesture to illustrate 'building up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [0.6523400545120239, 0.8232717514038086], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [0.7093172073364258, 0.8232717514038086], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005702.pt"}
{"idx": 5704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VMbC4Fww1P4.mp4", "duration": 85.83574999999999, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where red liquid is poured onto the underwear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where red liquid is poured onto the underwear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 50.0], "base_se_norm": [0.04851424694061279, 0.1650158166885376], "gt_se_raw": [45.0, 50.0], "gt_se_norm": [0.04851424694061279, 0.1650158166885376], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005704.pt"}
{"idx": 5706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VMbC4Fww1P4.mp4", "duration": 85.83574999999999, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the absorbency levels are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the absorbency levels are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 44.0], "base_se_norm": [-0.2543898820877075, 0.02521383762359619], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.2543898820877075, -0.11458802223205566], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005706.pt"}
{"idx": 5708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1qM0p24SNhc.mp4", "duration": 35.375, "prompt_text": "Please find the visual event described by the sentence 'When does the woman say she went skiing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman say she went skiing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.434628963470459, -0.3215547800064087], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.434628963470459, -0.37809187173843384], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005708.pt"}
{"idx": 5710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1qM0p24SNhc.mp4", "duration": 35.375, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman ask about the weather?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman ask about the weather?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 16.0], "base_se_norm": [-0.15194344520568848, -0.09540635347366333], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.15194344520568848, -0.038869261741638184], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005710.pt"}
{"idx": 5712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1qM0p24SNhc.mp4", "duration": 35.375, "prompt_text": "Please find the visual event described by the sentence 'When do the two people decide to go and walk off together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two people decide to go and walk off together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [0.5830389261245728, 0.7526501417160034], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [0.5830389261245728, 0.7526501417160034], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005712.pt"}
{"idx": 5714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8B8rjjbVApE.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the students are wearing lead vests and looking at a monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the students are wearing lead vests and looking at a monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.5414421558380127, -0.017376065254211426], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005714.pt"}
{"idx": 5716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8B8rjjbVApE.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a woman in a grey scrub top and blue gloves is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a woman in a grey scrub top and blue gloves is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3756735324859619, 0.8342313766479492], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005716.pt"}
{"idx": 5718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hc1hzSvynVY.mp4", "duration": 5.866666666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the woman and child engaged in a playful tug-of-war with a green object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the woman and child engaged in a playful tug-of-war with a green object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7045453786849976], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7045453786849976], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005718.pt"}
{"idx": 5720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hc1hzSvynVY.mp4", "duration": 5.866666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman in the purple top seen smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman in the purple top seen smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7045453786849976], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7045453786849976], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005720.pt"}
{"idx": 5722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hc1hzSvynVY.mp4", "duration": 5.866666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where two people are holding a green ring-shaped toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where two people are holding a green ring-shaped toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7045453786849976], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7045453786849976], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005722.pt"}
{"idx": 5724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ApN73TUVMEU.mp4", "duration": 145.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat flush the toilet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat flush the toilet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.35439562797546387, -0.23076927661895752], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.24450552463531494, -0.2032967209815979], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005724.pt"}
{"idx": 5726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ApN73TUVMEU.mp4", "duration": 145.6, "prompt_text": "Please find the visual event described by the sentence 'When does the cat turn on the faucet and cause the sink to overflow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat turn on the faucet and cause the sink to overflow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.2170330286026001, -0.13461542129516602], "gt_se_raw": [100.0, 106.0], "gt_se_norm": [0.37362635135650635, 0.4560438394546509], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005726.pt"}
{"idx": 5728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N_yBGQqg7kM.mp4", "duration": 148.89874999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the filmmaker and LGBT advocate state their pronoun is 'he'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the filmmaker and LGBT advocate state their pronoun is 'he'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [-0.8522485494613647, -0.7582249641418457], "gt_se_raw": [11.0, 19.0], "gt_se_norm": [-0.8522485494613647, -0.7447929978370667], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005728.pt"}
{"idx": 5730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N_yBGQqg7kM.mp4", "duration": 148.89874999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the leather jacket explain that some people use 'they' as a pronoun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the leather jacket explain that some people use 'they' as a pronoun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 100.0], "base_se_norm": [0.276034951210022, 0.34319472312927246], "gt_se_raw": [95.0, 101.0], "gt_se_norm": [0.276034951210022, 0.35662662982940674], "start_token_span": [77, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005730.pt"}
{"idx": 5732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N_yBGQqg7kM.mp4", "duration": 148.89874999999998, "prompt_text": "Please find the visual event described by the sentence 'When does a woman with braided hair and glasses talk about situations where gendered language isn't necessary?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman with braided hair and glasses talk about situations where gendered language isn't necessary?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 83.0], "base_se_norm": [-0.03289979696273804, 0.11485159397125244], "gt_se_raw": [72.0, 84.0], "gt_se_norm": [-0.03289979696273804, 0.12828350067138672], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005732.pt"}
{"idx": 5734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mxc-TvEhWec.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the boy is guided to touch the wall on his left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the boy is guided to touch the wall on his left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.8306778073310852, -0.26627039909362793], "gt_se_raw": [3.0, 14.0], "gt_se_norm": [-0.8306778073310852, -0.2098296880722046], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005734.pt"}
{"idx": 5736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mxc-TvEhWec.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman praise the boy by saying 'Good job'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman praise the boy by saying 'Good job'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.6367813348770142, 0.7496628761291504], "gt_se_raw": [14.0, 20.0], "gt_se_norm": [-0.2098296880722046, 0.128814697265625], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005736.pt"}
{"idx": 5738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pRQLM5nJIeE.mp4", "duration": 122.04, "prompt_text": "Please find the visual event described by the sentence 'When is the baby lying on the bed alone, laughing and moving his arms and legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the baby lying on the bed alone, laughing and moving his arms and legs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.934447705745697], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.934447705745697], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005738.pt"}
{"idx": 5740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pRQLM5nJIeE.mp4", "duration": 122.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the baby laugh and fall back onto the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the baby laugh and fall back onto the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.7377908825874329, -0.6886266469955444], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.16420847177505493, -0.1150442361831665], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005740.pt"}
{"idx": 5742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pRQLM5nJIeE.mp4", "duration": 122.04, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the baby is playing with a green stuffed toy dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the baby is playing with a green stuffed toy dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 121.0], "base_se_norm": [0.933792233467102, 0.9829564094543457], "gt_se_raw": [119.0, 122.0], "gt_se_norm": [0.9501802921295166, 0.9993444681167603], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005742.pt"}
{"idx": 5744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DpN1zNVoI9E.mp4", "duration": 14.429383471797113, "prompt_text": "Please find the visual event described by the sentence 'When does the wheel begin to decelerate, making the numbers more distinct?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the wheel begin to decelerate, making the numbers more distinct?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.30696970224380493, 0.38606059551239014], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.30696970224380493, 0.1088484525680542], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005744.pt"}
{"idx": 5746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DpN1zNVoI9E.mp4", "duration": 14.429383471797113, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the '1000' jackpot slot passes the pointer as the wheel slows down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the '1000' jackpot slot passes the pointer as the wheel slows down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.30696970224380493, 0.1088484525680542], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.16836363077163696, -0.029757559299468994], "start_token_span": [77, 80], "end_token_span": [82, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005746.pt"}
{"idx": 5748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/69XmzLpL_7A.mp4", "duration": 96.93016666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone washing their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone washing their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.21593040227890015, -0.13339674472808838], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.21593040227890015, -0.13339674472808838], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005748.pt"}
{"idx": 5750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/69XmzLpL_7A.mp4", "duration": 96.93016666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman shown shaking her finger to indicate 'no'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman shown shaking her finger to indicate 'no'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 59.0], "base_se_norm": [0.17610442638397217, 0.21737122535705566], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [0.17610442638397217, 0.21737122535705566], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005750.pt"}
{"idx": 5752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/69XmzLpL_7A.mp4", "duration": 96.93016666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the disposable mask is being thrown into a bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the disposable mask is being thrown into a bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [0.5475058555603027, 0.6094061136245728], "gt_se_raw": [75.0, 77.0], "gt_se_norm": [0.5475058555603027, 0.5887726545333862], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005752.pt"}
{"idx": 5754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tXKw_rBhOTs.mp4", "duration": 45.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the white cat ring the bell, then immediately get fed by a hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the white cat ring the bell, then immediately get fed by a hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 25.0], "base_se_norm": [-0.24500370025634766, 0.11028873920440674], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.06587719917297363, 0.15470027923583984], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005754.pt"}
{"idx": 5756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tXKw_rBhOTs.mp4", "duration": 45.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the last time a person's hand appears in the video to give a treat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the last time a person's hand appears in the video to give a treat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 35.0], "base_se_norm": [0.4655810594558716, 0.5544041395187378], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [0.6876387596130371, 0.7764619588851929], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005756.pt"}
{"idx": 5758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5cLisAxT_v8.mp4", "duration": 55.346958333333326, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention the global number of blind people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention the global number of blind people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.024336636066436768, 0.08407044410705566], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.7831858992576599, -0.6747788786888123], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005758.pt"}
{"idx": 5760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5cLisAxT_v8.mp4", "duration": 55.346958333333326, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man state that his team has spent six years on this technology?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man state that his team has spent six years on this technology?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.024336636066436768, 0.08407044410705566], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.16887933015823364, -0.024336636066436768], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005760.pt"}
{"idx": 5762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5cLisAxT_v8.mp4", "duration": 55.346958333333326, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for when the 3DPHOTOWORKS logo is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for when the 3DPHOTOWORKS logo is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [0.879055380821228, 0.9874624013900757], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [0.879055380821228, 0.9874624013900757], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005762.pt"}
{"idx": 5764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jQHVrDFNmJE.mp4", "duration": 125.75909242575909, "prompt_text": "Please find the visual event described by the sentence 'When does the video display a technical diagram about how smart textiles work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display a technical diagram about how smart textiles work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.7455452680587769, -0.6501246690750122], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [-0.7455452680587769, -0.6501246690750122], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005764.pt"}
{"idx": 5766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jQHVrDFNmJE.mp4", "duration": 125.75909242575909, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when a man and a woman are collaborating on a piece of fabric.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when a man and a woman are collaborating on a piece of fabric.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [-0.31615281105041504, -0.25253915786743164], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.31615281105041504, -0.25253915786743164], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005766.pt"}
{"idx": 5768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFy3jZk1Tko.mp4", "duration": 68.8, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the title card 'Simon's Cat PRESENTS UFO' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the title card 'Simon's Cat PRESENTS UFO' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.854651153087616, -0.7674418687820435], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.8837209343910217, -0.7965116500854492], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005768.pt"}
{"idx": 5770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFy3jZk1Tko.mp4", "duration": 68.8, "prompt_text": "Please find the visual event described by the sentence 'At what point does the butterfly land on the cat's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the butterfly land on the cat's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.2732558250427246, -0.12790703773498535], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [-0.30232560634613037, -0.1860465407371521], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005770.pt"}
{"idx": 5772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFy3jZk1Tko.mp4", "duration": 68.8, "prompt_text": "Please find the visual event described by the sentence 'When does the camera pan out to reveal the crop circles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera pan out to reveal the crop circles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [0.25, 0.39534878730773926], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [0.25, 0.3662790060043335], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005772.pt"}
{"idx": 5774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6PXKSiJchU.mp4", "duration": 113.15470833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the flour added to the butter to make the sauce base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the flour added to the butter to make the sauce base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [-0.20462876558303833, -0.15160399675369263], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [-0.20462876558303833, -0.15160399675369263], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005774.pt"}
{"idx": 5776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6PXKSiJchU.mp4", "duration": 113.15470833333333, "prompt_text": "Please find the visual event described by the sentence 'When are all the ingredients mixed together in the baking dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the ingredients mixed together in the baking dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [0.36096858978271484, 0.44934308528900146], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [0.36096858978271484, 0.43166816234588623], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005776.pt"}
{"idx": 5778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wezZVZXFO3U.mp4", "duration": 158.3, "prompt_text": "Please find the visual event described by the sentence 'When does the woman demonstrate lining the baking pan with foil?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman demonstrate lining the baking pan with foil?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 33.0], "base_se_norm": [-0.7599494457244873, -0.5830701589584351], "gt_se_raw": [16.0, 34.0], "gt_se_norm": [-0.7978521585464478, -0.5704358816146851], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005778.pt"}
{"idx": 5780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wezZVZXFO3U.mp4", "duration": 158.3, "prompt_text": "Please find the visual event described by the sentence 'At what point are the graham cracker crumbs poured into the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the graham cracker crumbs poured into the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 62.0], "base_se_norm": [-0.25457990169525146, -0.21667718887329102], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.25457990169525146, -0.2040429711341858], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005780.pt"}
{"idx": 5782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wezZVZXFO3U.mp4", "duration": 158.3, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is adding and mixing the eggs into the batter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is adding and mixing the eggs into the batter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 116.0], "base_se_norm": [0.23815536499023438, 0.46557164192199707], "gt_se_raw": [98.0, 116.0], "gt_se_norm": [0.23815536499023438, 0.46557164192199707], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005782.pt"}
{"idx": 5784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0NPYZyI7V8.mp4", "duration": 127.21041666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man pulls the wide end of the tie underneath to the other side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man pulls the wide end of the tie underneath to the other side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 67.0], "base_se_norm": [-0.18245691061019897, 0.053372859954833984], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.056681036949157715, -0.009515106678009033], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005784.pt"}
{"idx": 5786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0NPYZyI7V8.mp4", "duration": 127.21041666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the time when he tightens the knot up to his collar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when he tightens the knot up to his collar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 117.0], "base_se_norm": [0.6508082151412964, 0.8394719362258911], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.5879203081130981, 0.6193642616271973], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005786.pt"}
{"idx": 5788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Alc7AhymB9o.mp4", "duration": 45.71233333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the website being refreshed but still showing the maintenance page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the website being refreshed but still showing the maintenance page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 28.0], "base_se_norm": [-0.38747382164001465, 0.2250523567199707], "gt_se_raw": [15.0, 28.0], "gt_se_norm": [-0.34372198581695557, 0.2250523567199707], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005788.pt"}
{"idx": 5790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Alc7AhymB9o.mp4", "duration": 45.71233333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the refreshed NYU website with the city skyline successfully load?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the refreshed NYU website with the city skyline successfully load?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [0.2688041925430298, 0.5313154458999634], "gt_se_raw": [29.0, 35.0], "gt_se_norm": [0.2688041925430298, 0.5313154458999634], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005790.pt"}
{"idx": 5792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Alc7AhymB9o.mp4", "duration": 45.71233333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a message states 'The NYU Website will Be Back Soon'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a message states 'The NYU Website will Be Back Soon'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 28.0], "base_se_norm": [-0.38747382164001465, 0.2250523567199707], "gt_se_raw": [13.0, 28.0], "gt_se_norm": [-0.4312257170677185, 0.2250523567199707], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005792.pt"}
{"idx": 5794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OA1ZRGFKRVI.mp4", "duration": 57.09870833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment people are making DIY coffee face scrub.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment people are making DIY coffee face scrub.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.05426931381225586, 0.05081188678741455], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.08929640054702759, 0.05081188678741455], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005794.pt"}
{"idx": 5796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OA1ZRGFKRVI.mp4", "duration": 57.09870833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is there a close-up of Lavido brand skincare products on a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there a close-up of Lavido brand skincare products on a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.7197835445404053, -0.6497293710708618], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.7197835445404053, -0.6497293710708618], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005796.pt"}
{"idx": 5798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wx-kvJlKvJI.mp4", "duration": 90.08999999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see two football players doing push-ups on the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two football players doing push-ups on the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 40.0], "base_se_norm": [-0.15639913082122803, -0.11199909448623657], "gt_se_raw": [39.0, 40.0], "gt_se_norm": [-0.1341990828514099, -0.11199909448623657], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005798.pt"}
{"idx": 5800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wx-kvJlKvJI.mp4", "duration": 90.08999999999999, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the scene where a man is running on a cobblestone street at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the scene where a man is running on a cobblestone street at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 56.0], "base_se_norm": [0.17660117149353027, 0.24320125579833984], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [0.17660117149353027, 0.2210012674331665], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005800.pt"}
{"idx": 5802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wx-kvJlKvJI.mp4", "duration": 90.08999999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a pink shirt spike a volleyball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a pink shirt spike a volleyball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 76.0], "base_se_norm": [0.6650017499923706, 0.687201738357544], "gt_se_raw": [74.0, 75.0], "gt_se_norm": [0.6428017616271973, 0.6650017499923706], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005802.pt"}
{"idx": 5804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o_P3sKbWbWQ.mp4", "duration": 166.4, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the 'Connection Lost' screen is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the 'Connection Lost' screen is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 97.0], "base_se_norm": [0.11778855323791504, 0.16586542129516602], "gt_se_raw": [92.0, 97.0], "gt_se_norm": [0.10576927661895752, 0.16586542129516602], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005804.pt"}
{"idx": 5806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o_P3sKbWbWQ.mp4", "duration": 166.4, "prompt_text": "Please find the visual event described by the sentence 'When does the character with a panda on their shirt get attacked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character with a panda on their shirt get attacked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 126.0], "base_se_norm": [0.46634626388549805, 0.514423131942749], "gt_se_raw": [124.0, 126.0], "gt_se_norm": [0.49038469791412354, 0.514423131942749], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005806.pt"}
{"idx": 5808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G6fMV1UPzkg.mp4", "duration": 147.04704704704704, "prompt_text": "Please find the visual event described by the sentence 'When are the robot dogs trotting around a soccer ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the robot dogs trotting around a soccer ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 58.0], "base_se_norm": [-0.7279782295227051, -0.2111368179321289], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.7007760405540466, -0.6191694736480713], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005808.pt"}
{"idx": 5810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G6fMV1UPzkg.mp4", "duration": 147.04704704704704, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the robot dogs perform a synchronized backflip?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the robot dogs perform a synchronized backflip?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 124.0], "base_se_norm": [0.6321307420730591, 0.686535120010376], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [0.033682823181152344, 0.060884952545166016], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005810.pt"}
{"idx": 5812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G6fMV1UPzkg.mp4", "duration": 147.04704704704704, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the robots are lined up in a row and simultaneously perform a bowing motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the robots are lined up in a row and simultaneously perform a bowing motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 75.0], "base_se_norm": [-0.10232806205749512, 0.020081758499145508], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.10232806205749512, -0.03432261943817139], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005812.pt"}
{"idx": 5814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W3v7-3y5THk.mp4", "duration": 178.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the instructor place a stack of sealed packages on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the instructor place a stack of sealed packages on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.75331711769104, -0.6860399842262268], "gt_se_raw": [24.0, 29.0], "gt_se_norm": [-0.7308914661407471, -0.6748270988464355], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005814.pt"}
{"idx": 5816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W3v7-3y5THk.mp4", "duration": 178.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the instructor helps the woman peel the backing from a sticker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the instructor helps the woman peel the backing from a sticker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 63.0], "base_se_norm": [-0.4393571615219116, -0.2935900092124939], "gt_se_raw": [92.0, 96.0], "gt_se_norm": [0.03158283233642578, 0.07643425464630127], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005816.pt"}
{"idx": 5818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bPa0TiCuvR4.mp4", "duration": 143.30983333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the matchmaker call out Mulan's name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the matchmaker call out Mulan's name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9860442280769348, -0.9302211403846741], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9860442280769348, -0.9720884561538696], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005818.pt"}
{"idx": 5820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bPa0TiCuvR4.mp4", "duration": 143.30983333333333, "prompt_text": "Please find the visual event described by the sentence 'When does Mulan start pouring tea for the matchmaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Mulan start pouring tea for the matchmaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 78.0], "base_se_norm": [0.018771767616271973, 0.08855056762695312], "gt_se_raw": [66.0, 82.0], "gt_se_norm": [-0.07891875505447388, 0.14437365531921387], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005820.pt"}
{"idx": 5822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bPa0TiCuvR4.mp4", "duration": 143.30983333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the matchmaker declare that Mulan will never bring her family honor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the matchmaker declare that Mulan will never bring her family honor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9860442280769348, -0.8604422211647034], "gt_se_raw": [133.0, 137.0], "gt_se_norm": [0.8561183214187622, 0.911941409111023], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005822.pt"}
{"idx": 5824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uasqlGvQhaM.mp4", "duration": 29.4, "prompt_text": "Please find the visual event described by the sentence 'When do the women clap and cheer for the boy's success?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the women clap and cheer for the boy's success?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [-0.2517006993293762, 0.22448980808258057], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.2517006993293762, -0.0476190447807312], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005824.pt"}
{"idx": 5826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uasqlGvQhaM.mp4", "duration": 29.4, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman in the black top gives a thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman in the black top gives a thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [0.22448980808258057, 0.36054420471191406], "gt_se_raw": [18.0, 19.0], "gt_se_norm": [0.22448980808258057, 0.2925170660018921], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005826.pt"}
{"idx": 5828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8KveZMqZU6M.mp4", "duration": 149.0, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the leopard-print shirt get flipped over the bar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the leopard-print shirt get flipped over the bar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.9328858852386475, -0.8926174640655518], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.9328858852386475, -0.8926174640655518], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005828.pt"}
{"idx": 5830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8KveZMqZU6M.mp4", "duration": 149.0, "prompt_text": "Please find the visual event described by the sentence 'At what point do the two men engage in a fight with pole weapons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the two men engage in a fight with pole weapons?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 41.0], "base_se_norm": [-0.5838925838470459, -0.4496644139289856], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [-0.5973154306411743, -0.5302013158798218], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005830.pt"}
{"idx": 5832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8KveZMqZU6M.mp4", "duration": 149.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the vampire showing his fangs with blood on his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the vampire showing his fangs with blood on his mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 81.0], "base_se_norm": [-0.03355705738067627, 0.08724832534790039], "gt_se_raw": [73.0, 75.0], "gt_se_norm": [-0.02013421058654785, 0.006711363792419434], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005832.pt"}
{"idx": 5834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Cdpf1Dl5b_4.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man is struggling to lift a barbell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man is struggling to lift a barbell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.20079922676086426, -0.0009990334510803223], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.20079922676086426, -0.06759911775588989], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005834.pt"}
{"idx": 5836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Cdpf1Dl5b_4.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man falls on the grass during a trust fall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man falls on the grass during a trust fall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [0.2654012441635132, 0.3986014127731323], "gt_se_raw": [20.0, 21.0], "gt_se_norm": [0.33200132846832275, 0.3986014127731323], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005836.pt"}
{"idx": 5838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CPf5cwGYsWs.mp4", "duration": 149.2, "prompt_text": "Please find the visual event described by the sentence 'When is the man pouring milk into his cereal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man pouring milk into his cereal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [-0.664879322052002, -0.5308310985565186], "gt_se_raw": [26.0, 35.0], "gt_se_norm": [-0.6514745354652405, -0.5308310985565186], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005838.pt"}
{"idx": 5840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CPf5cwGYsWs.mp4", "duration": 149.2, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man try to find and press the buttons in the elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man try to find and press the buttons in the elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 80.0], "base_se_norm": [-0.07506698369979858, 0.07238602638244629], "gt_se_raw": [70.0, 80.0], "gt_se_norm": [-0.06166219711303711, 0.07238602638244629], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005840.pt"}
{"idx": 5842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CPf5cwGYsWs.mp4", "duration": 149.2, "prompt_text": "Please find the visual event described by the sentence 'When is the man, guided by a woman, walking across a crosswalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man, guided by a woman, walking across a crosswalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 120.0], "base_se_norm": [0.5817694664001465, 0.608579158782959], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [0.5817694664001465, 0.6219840049743652], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005842.pt"}
{"idx": 5844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q3wJ32w4s_A.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman is looking at an X-ray of a pelvis on a monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman is looking at an X-ray of a pelvis on a monitor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 15.0], "gt_se_norm": [-0.4753605127334595, -0.016300976276397705], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005844.pt"}
{"idx": 5846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q3wJ32w4s_A.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the text 'LENGTH of PROGRAM' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the text 'LENGTH of PROGRAM' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.016300976276397705, 0.44275856018066406], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005846.pt"}
{"idx": 5848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9C6gDS9YGAU.mp4", "duration": 140.6405, "prompt_text": "Please find the visual event described by the sentence 'When is a player seen talking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a player seen talking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.6871455907821655, -0.6444836258888245], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.6871455907821655, -0.6587042808532715], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005848.pt"}
{"idx": 5850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9C6gDS9YGAU.mp4", "duration": 140.6405, "prompt_text": "Please find the visual event described by the sentence 'When does the player in the white jersey high-five his teammate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in the white jersey high-five his teammate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 60.0], "base_se_norm": [-0.17520201206207275, -0.1467607021331787], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.17520201206207275, -0.1467607021331787], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005850.pt"}
{"idx": 5852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9C6gDS9YGAU.mp4", "duration": 140.6405, "prompt_text": "Please find the visual event described by the sentence 'At what point do the players celebrate victory together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the players celebrate victory together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 122.0], "base_se_norm": [0.6922578811645508, 0.7349199056625366], "gt_se_raw": [119.0, 122.0], "gt_se_norm": [0.6922578811645508, 0.7349199056625366], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005852.pt"}
{"idx": 5854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u0p_dBCEDs4.mp4", "duration": 140.30683333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where spaghetti is poured into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where spaghetti is poured into the pot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.2730218768119812, -0.21600401401519775], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [-0.25876742601394653, -0.2017495036125183], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005854.pt"}
{"idx": 5856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u0p_dBCEDs4.mp4", "duration": 140.30683333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when tomato sauce is added to the cooked spaghetti.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when tomato sauce is added to the cooked spaghetti.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 124.0], "base_se_norm": [0.7247911691665649, 0.7675546407699585], "gt_se_raw": [121.0, 124.0], "gt_se_norm": [0.7247911691665649, 0.7675546407699585], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005856.pt"}
{"idx": 5858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_U-J9PqgmIc.mp4", "duration": 141.07426666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the teacher and the child holding hands and swaying to the music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the teacher and the child holding hands and swaying to the music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 85.0], "base_se_norm": [0.049092888832092285, 0.20503902435302734], "gt_se_raw": [74.0, 81.0], "gt_se_norm": [0.049092888832092285, 0.1483314037322998], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005858.pt"}
{"idx": 5860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_U-J9PqgmIc.mp4", "duration": 141.07426666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child verbally refuse the toothbrush by saying 'no thank you'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child verbally refuse the toothbrush by saying 'no thank you'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6172229051589966, -0.5746921300888062], "gt_se_raw": [109.0, 112.0], "gt_se_norm": [0.5452853441238403, 0.5878162384033203], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005860.pt"}
{"idx": 5862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_U-J9PqgmIc.mp4", "duration": 141.07426666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the child is dancing by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the child is dancing by himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 73.0], "base_se_norm": [-0.09267646074295044, 0.034915924072265625], "gt_se_raw": [63.0, 73.0], "gt_se_norm": [-0.1068534255027771, 0.034915924072265625], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005862.pt"}
{"idx": 5864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dZiJkicepzM.mp4", "duration": 160.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a cat is playing with a feather toy on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a cat is playing with a feather toy on the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 49.0], "base_se_norm": [-0.6755407452583313, -0.3885191082954407], "gt_se_raw": [26.0, 41.0], "gt_se_norm": [-0.6755407452583313, -0.4883527159690857], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005864.pt"}
{"idx": 5866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dZiJkicepzM.mp4", "duration": 160.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is the cat's bandaged leg shown by the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the cat's bandaged leg shown by the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 25.0], "base_se_norm": [-0.7628951668739319, -0.6880199909210205], "gt_se_raw": [19.0, 25.0], "gt_se_norm": [-0.7628951668739319, -0.6880199909210205], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005866.pt"}
{"idx": 5868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zW39WTnHCc8.mp4", "duration": 82.16541666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man have a shocked and wide-eyed expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man have a shocked and wide-eyed expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [0.4361274242401123, 0.5821743011474609], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [0.41178619861602783, 0.5821743011474609], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005868.pt"}
{"idx": 5870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zW39WTnHCc8.mp4", "duration": 82.16541666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man tilt his blue mug downwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man tilt his blue mug downwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [-0.07503658533096313, 0.0466691255569458], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [0.0466691255569458, 0.0953514575958252], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005870.pt"}
{"idx": 5872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zW39WTnHCc8.mp4", "duration": 82.16541666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the man standing still, holding his briefcase and mug, looking ready for work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man standing still, holding his briefcase and mug, looking ready for work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 28.0], "base_se_norm": [-1.0, -0.318448007106781], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, -0.5131771564483643], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005872.pt"}
{"idx": 5874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZsnaOQwOlI.mp4", "duration": 236.02745833333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman trying to make a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman trying to make a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 22.0], "base_se_norm": [-0.9237377047538757, -0.8135809898376465], "gt_se_raw": [9.0, 22.0], "gt_se_norm": [-0.9237377047538757, -0.8135809898376465], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005874.pt"}
{"idx": 5876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZsnaOQwOlI.mp4", "duration": 236.02745833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is destroying the lights in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is destroying the lights in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 163.0], "base_se_norm": [0.10156667232513428, 0.38119518756866455], "gt_se_raw": [121.0, 156.0], "gt_se_norm": [0.025304436683654785, 0.3218801021575928], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005876.pt"}
{"idx": 5878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSTNyHkde08.mp4", "duration": 297.422125, "prompt_text": "Please find the visual event described by the sentence 'A woman is shown using her laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is shown using her laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.9596533179283142, -0.9394799470901489], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.9529288411140442, -0.9327555298805237], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005878.pt"}
{"idx": 5880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSTNyHkde08.mp4", "duration": 297.422125, "prompt_text": "Please find the visual event described by the sentence 'A beetle is rolling a large ball of dung.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A beetle is rolling a large ball of dung.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 31.0], "base_se_norm": [-0.8453376293182373, -0.7915420532226562], "gt_se_raw": [83.0, 87.0], "gt_se_norm": [-0.44187068939208984, -0.4149729013442993], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005880.pt"}
{"idx": 5882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSTNyHkde08.mp4", "duration": 297.422125, "prompt_text": "Please find the visual event described by the sentence 'Two beetles with horns are shown fighting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Two beetles with horns are shown fighting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 125.0], "base_se_norm": [-0.19979053735733032, -0.15944379568099976], "gt_se_raw": [120.0, 122.0], "gt_se_norm": [-0.1930660605430603, -0.17961716651916504], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005882.pt"}
{"idx": 5884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZykGz5kF0U.mp4", "duration": 292.04175, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the orange gets extra cheese for its pizza.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the orange gets extra cheese for its pizza.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 53.0], "base_se_norm": [-0.7055215835571289, -0.6370382308959961], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [-0.7192182540893555, -0.6918249130249023], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005884.pt"}
{"idx": 5886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZykGz5kF0U.mp4", "duration": 292.04175, "prompt_text": "Please find the visual event described by the sentence 'At what point does the orange receive the pepperoni topping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the orange receive the pepperoni topping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 93.0], "base_se_norm": [-0.41104310750961304, -0.36310476064682007], "gt_se_raw": [86.0, 88.0], "gt_se_norm": [-0.41104310750961304, -0.3973464369773865], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005886.pt"}
{"idx": 5888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sL3vqlHabck.mp4", "duration": 171.00416666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy, wearing a striped shirt, first appear on the green sofa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy, wearing a striped shirt, first appear on the green sofa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [-0.8713481426239014, -0.7894787788391113], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.8713481426239014, -0.8362612724304199], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005888.pt"}
{"idx": 5890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sL3vqlHabck.mp4", "duration": 171.00416666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy say 'Park, okay'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy say 'Park, okay'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 71.0], "base_se_norm": [-0.2163933515548706, -0.169610857963562], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.39182770252227783, -0.36843645572662354], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005890.pt"}
{"idx": 5892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sL3vqlHabck.mp4", "duration": 171.00416666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the girl is pressing a button on the phone after the call.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 158.0 - 166.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the girl is pressing a button on the phone after the call.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 158.0 - 166.0 seconds.<|im_end|>\n", "base_se_raw": [158.0, 166.0], "base_se_norm": [0.8479082584381104, 0.941473126411438], "gt_se_raw": [160.0, 164.0], "gt_se_norm": [0.8712993860244751, 0.9180818796157837], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005892.pt"}
{"idx": 5894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJTgEAuCI8Y.mp4", "duration": 59.684625, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the team is celebrating with confetti falling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the team is celebrating with confetti falling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 35.0], "base_se_norm": [-0.09524434804916382, 0.17283141613006592], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [-0.09524434804916382, 0.1393219232559204], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005894.pt"}
{"idx": 5896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJTgEAuCI8Y.mp4", "duration": 59.684625, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range where a woman is holding a helmet up high at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range where a woman is holding a helmet up high at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9329810738563538], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9329810738563538], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005896.pt"}
{"idx": 5898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FE6BiUi5HsU.mp4", "duration": 127.7276, "prompt_text": "Please find the visual event described by the sentence 'When can we see a player in a black suit with a TNT block on his head jumping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a player in a black suit with a TNT block on his head jumping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 106.0], "base_se_norm": [0.6128072738647461, 0.6597821712493896], "gt_se_raw": [105.0, 106.0], "gt_se_norm": [0.644123911857605, 0.6597821712493896], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005898.pt"}
{"idx": 5900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FE6BiUi5HsU.mp4", "duration": 127.7276, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when a player with purple hair is seen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when a player with purple hair is seen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 112.0], "base_se_norm": [0.5814906358718872, 0.7537322044372559], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.5814906358718872, 0.6128072738647461], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005900.pt"}
{"idx": 5902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FE6BiUi5HsU.mp4", "duration": 127.7276, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the message 'The TNT has been released!' is displayed in the chat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the message 'The TNT has been released!' is displayed in the chat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [0.17437422275543213, 0.22134923934936523], "gt_se_raw": [74.0, 76.0], "gt_se_norm": [0.1587158441543579, 0.1900324821472168], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005902.pt"}
{"idx": 5904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2j6PKrsoy6U.mp4", "duration": 11.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy wakes up with an angry expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy wakes up with an angry expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.8250728845596313, -0.47521865367889404], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.8250728845596313, -0.47521865367889404], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005904.pt"}
{"idx": 5906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2j6PKrsoy6U.mp4", "duration": 11.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the boy is screaming in frustration.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the boy is screaming in frustration.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 9.0], "base_se_norm": [-0.6501457691192627, 0.5743440389633179], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.3994169235229492, 0.5743440389633179], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005906.pt"}
{"idx": 5908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AURiyMmwtOU.mp4", "duration": 319.5192, "prompt_text": "Please find the visual event described by the sentence 'Where can we see people disembarking from a boat onto a dock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see people disembarking from a boat onto a dock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 91.0], "base_se_norm": [-0.48046940565109253, -0.43039417266845703], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [-0.48046940565109253, -0.44917237758636475], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005908.pt"}
{"idx": 5910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AURiyMmwtOU.mp4", "duration": 319.5192, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where several people are kayaking into a sea cave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 278.0 - 288.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where several people are kayaking into a sea cave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 278.0 - 288.0 seconds.<|im_end|>\n", "base_se_raw": [278.0, 288.0], "base_se_norm": [0.7401145696640015, 0.802708625793457], "gt_se_raw": [279.0, 285.0], "gt_se_norm": [0.7463738918304443, 0.7839304208755493], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005910.pt"}
{"idx": 5912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AURiyMmwtOU.mp4", "duration": 319.5192, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a large group of seals resting on a sandy beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 156.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a large group of seals resting on a sandy beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 156.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [156.0, 160.0], "base_se_norm": [-0.023532867431640625, 0.0015047788619995117], "gt_se_raw": [156.0, 158.0], "gt_se_norm": [-0.023532867431640625, -0.011014044284820557], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005912.pt"}
{"idx": 5914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UPBMG5EYydo.mp4", "duration": 250.8, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for explaining the elements necessary for photosynthesis, such as sunlight and carbon dioxide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for explaining the elements necessary for photosynthesis, such as sunlight and carbon dioxide?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 59.0], "base_se_norm": [-0.6889952421188354, -0.5295056104660034], "gt_se_raw": [39.0, 58.0], "gt_se_norm": [-0.6889952421188354, -0.5374800562858582], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005914.pt"}
{"idx": 5916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UPBMG5EYydo.mp4", "duration": 250.8, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the food chain pyramid is displayed and explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 172.0 - 177.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the food chain pyramid is displayed and explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 172.0 - 177.0 seconds.<|im_end|>\n", "base_se_raw": [172.0, 177.0], "base_se_norm": [0.3716108798980713, 0.4114832878112793], "gt_se_raw": [172.0, 177.0], "gt_se_norm": [0.3716108798980713, 0.4114832878112793], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005916.pt"}
{"idx": 5918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q_P8WCbhC6s.mp4", "duration": 118.743625, "prompt_text": "Please find the visual event described by the sentence 'At what point does a hand in a black glove take a red phone from a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a hand in a black glove take a red phone from a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [-0.1410064697265625, -0.09047746658325195], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.1410064697265625, -0.10732048749923706], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005918.pt"}
{"idx": 5920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q_P8WCbhC6s.mp4", "duration": 118.743625, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the masked man appears outside the glass door and shows the woman her picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the masked man appears outside the glass door and shows the woman her picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 88.0], "base_se_norm": [0.3979698419570923, 0.48218488693237305], "gt_se_raw": [81.0, 89.0], "gt_se_norm": [0.36428380012512207, 0.4990278482437134], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005920.pt"}
{"idx": 5922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q_P8WCbhC6s.mp4", "duration": 118.743625, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the woman tries to climb up to the roof.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the woman tries to climb up to the roof.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 96.0], "base_se_norm": [0.5495569705963135, 0.6169289350509644], "gt_se_raw": [100.0, 102.0], "gt_se_norm": [0.6843010187149048, 0.717987060546875], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005922.pt"}
{"idx": 5924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fWDaRN490BI.mp4", "duration": 166.3662, "prompt_text": "Please find the visual event described by the sentence 'At what point is a coffin shown strapped to a wheelchair during a protest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a coffin shown strapped to a wheelchair during a protest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 72.0], "base_se_norm": [-0.17050457000732422, -0.13443952798843384], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [-0.17050457000732422, -0.13443952798843384], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005924.pt"}
{"idx": 5926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fWDaRN490BI.mp4", "duration": 166.3662, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip of the child saying, \"You are our inspiration.\"', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip of the child saying, \"You are our inspiration.\"', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 92.0], "base_se_norm": [0.08195054531097412, 0.10599386692047119], "gt_se_raw": [91.0, 92.0], "gt_se_norm": [0.09397220611572266, 0.10599386692047119], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005926.pt"}
{"idx": 5928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kKhZtOzakXA.mp4", "duration": 107.39895833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the red jacket first appear and start speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the red jacket first appear and start speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.925511360168457, -0.7579119801521301], "gt_se_raw": [5.0, 31.0], "gt_se_norm": [-0.9068892002105713, -0.4227132201194763], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005928.pt"}
{"idx": 5930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kKhZtOzakXA.mp4", "duration": 107.39895833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman sits on the grass with a 'YOU HAVE ARRIVED' banner in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman sits on the grass with a 'YOU HAVE ARRIVED' banner in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.25511378049850464, -0.1992473602294922], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.2364916205406189, -0.1992473602294922], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005930.pt"}
{"idx": 5932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kKhZtOzakXA.mp4", "duration": 107.39895833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a man in a green polo shirt talking to the camera while others pose for a photo behind him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a man in a green polo shirt talking to the camera while others pose for a photo behind him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 80.0], "base_se_norm": [0.3407951593399048, 0.4897724390029907], "gt_se_raw": [73.0, 81.0], "gt_se_norm": [0.3594173192977905, 0.5083945989608765], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005932.pt"}
{"idx": 5934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X4hPwKhjJk0.mp4", "duration": 164.03053333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man is continuously stirring the mixture in the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man is continuously stirring the mixture in the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 71.0], "base_se_norm": [-0.7439501285552979, -0.13430750370025635], "gt_se_raw": [21.0, 32.0], "gt_se_norm": [-0.7439501285552979, -0.6098287105560303], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005934.pt"}
{"idx": 5936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X4hPwKhjJk0.mp4", "duration": 164.03053333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man grating a lime over the glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 146.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man grating a lime over the glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 146.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [146.0, 156.0], "base_se_norm": [0.7801563739776611, 0.9020849466323853], "gt_se_raw": [146.0, 157.0], "gt_se_norm": [0.7801563739776611, 0.9142777919769287], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005936.pt"}
{"idx": 5938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V9WcxlXAWCM.mp4", "duration": 31.598333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman in a white t-shirt talking to a group of people in a kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman in a white t-shirt talking to a group of people in a kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.7468221187591553, -0.49364417791366577], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.7468221187591553, -0.6202331781387329], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005938.pt"}
{"idx": 5940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V9WcxlXAWCM.mp4", "duration": 31.598333333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a woman in a plaid shirt is cutting strawberries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a woman in a plaid shirt is cutting strawberries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [0.13930058479309082, 0.32918405532836914], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [0.20259511470794678, 0.32918405532836914], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005940.pt"}
{"idx": 5942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V9WcxlXAWCM.mp4", "duration": 31.598333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are people sitting at a table and eating their meal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people sitting at a table and eating their meal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [0.7722454071044922, 0.962128758430481], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [0.8355398178100586, 0.962128758430481], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005942.pt"}
{"idx": 5944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_bdAJXsjvk.mp4", "duration": 8.141466666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when the fluffy white bird is making a hissing sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when the fluffy white bird is making a hissing sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, 0.7195917367935181], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.9652477502822876], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005944.pt"}
{"idx": 5946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_bdAJXsjvk.mp4", "duration": 8.141466666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the bird is shown with its beak visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the bird is shown with its beak visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, 0.7195917367935181], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.9652477502822876], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005946.pt"}
{"idx": 5948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZxcXr7WaGvc.mp4", "duration": 98.80704166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a young boy receives a large check.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a young boy receives a large check.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.8380682468414307, -0.7975852489471436], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [-0.8178267478942871, -0.7975852489471436], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005948.pt"}
{"idx": 5950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZxcXr7WaGvc.mp4", "duration": 98.80704166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a person is using a handheld electronic magnifier on a newspaper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a person is using a handheld electronic magnifier on a newspaper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [0.03231501579284668, 0.07279801368713379], "gt_se_raw": [51.0, 52.0], "gt_se_norm": [0.03231501579284668, 0.052556514739990234], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005950.pt"}
{"idx": 5952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZxcXr7WaGvc.mp4", "duration": 98.80704166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a group of children in green shirts cheering around a bronze statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a group of children in green shirts cheering around a bronze statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 82.0], "base_se_norm": [0.6193177700042725, 0.65980064868927], "gt_se_raw": [81.0, 83.0], "gt_se_norm": [0.6395591497421265, 0.6800421476364136], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005952.pt"}
{"idx": 5954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/czTG7JxOruo.mp4", "duration": 151.38456666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the children gathered around the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children gathered around the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.5772356390953064, -0.5243901014328003], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.5772356390953064, -0.5243901014328003], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005954.pt"}
{"idx": 5956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/czTG7JxOruo.mp4", "duration": 151.38456666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a screen recording demonstrates auditing a website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a screen recording demonstrates auditing a website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 86.0], "base_se_norm": [0.004065394401550293, 0.13617920875549316], "gt_se_raw": [76.0, 83.0], "gt_se_norm": [0.004065394401550293, 0.09654510021209717], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005956.pt"}
{"idx": 5958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4qMyGXTcsg.mp4", "duration": 31.298, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman helps the boy draw on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman helps the boy draw on the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.6165889501571655], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.8082944750785828, -0.5526870489120483], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005958.pt"}
{"idx": 5960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4qMyGXTcsg.mp4", "duration": 31.298, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman touches the boy's nose and mouth area with her fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman touches the boy's nose and mouth area with her fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 28.0], "base_se_norm": [0.15023326873779297, 0.7892516851425171], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [0.15023326873779297, 0.34193873405456543], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005960.pt"}
{"idx": 5962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4qMyGXTcsg.mp4", "duration": 31.298, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman briefly smile at the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman briefly smile at the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [0.5975461006164551, 0.7253499031066895], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [0.6614480018615723, 0.7253499031066895], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005962.pt"}
{"idx": 5964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_CE8-CTcRPs.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'A female instructor is teaching a group of female students how to operate a ventilator machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A female instructor is teaching a group of female students how to operate a ventilator machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.5737595558166504, -0.2540792226791382], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.6270396113395691, -0.20079916715621948], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005964.pt"}
{"idx": 5966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_CE8-CTcRPs.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the text 'LENGTH of PROGRAM' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the text 'LENGTH of PROGRAM' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.17216122150421143, 0.49184155464172363], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.17216122150421143, 0.49184155464172363], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005966.pt"}
{"idx": 5968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OkBaZLq7gnU.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman picking up a bottle of perfume?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman picking up a bottle of perfume?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8688042163848877, -0.829445481300354], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [-0.829445481300354, -0.8163259625434875], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005968.pt"}
{"idx": 5970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OkBaZLq7gnU.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the sales associate hand the woman a shopping bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the sales associate hand the woman a shopping bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [-0.31778204441070557, -0.2784233093261719], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.31778204441070557, -0.2915428876876831], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005970.pt"}
{"idx": 5972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OkBaZLq7gnU.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman take items from a golden luggage cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman take items from a golden luggage cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 113.0], "base_se_norm": [0.31195759773254395, 0.4825119972229004], "gt_se_raw": [104.0, 107.0], "gt_se_norm": [0.36443591117858887, 0.40379464626312256], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005972.pt"}
{"idx": 5974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5SDeLotL7u0.mp4", "duration": 30.1301, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a family is walking together on a path in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a family is walking together on a path in the woods.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.668105959892273, -0.46896958351135254], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.7344847917556763, -0.5353484153747559], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005974.pt"}
{"idx": 5976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5SDeLotL7u0.mp4", "duration": 30.1301, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where a family is moving into a new house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where a family is moving into a new house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.07069677114486694, 0.26119720935821533], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.07069677114486694, 0.12843966484069824], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005976.pt"}
{"idx": 5978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fF8cAIfSbMw.mp4", "duration": 45.1451, "prompt_text": "Please find the visual event described by the sentence 'When can we see the adult massaging the boy's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the adult massaging the boy's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 22.0], "base_se_norm": [-0.9556984305381775, -0.02536487579345703], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9556984305381775, -0.3354760408401489], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005978.pt"}
{"idx": 5980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fF8cAIfSbMw.mp4", "duration": 45.1451, "prompt_text": "Please find the visual event described by the sentence 'At what point does the adult hold both of the boy's hands together and rub them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the adult hold both of the boy's hands together and rub them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9556984305381775, -0.4240792393684387], "gt_se_raw": [19.0, 26.0], "gt_se_norm": [-0.15826964378356934, 0.15184152126312256], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005980.pt"}
{"idx": 5982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fF8cAIfSbMw.mp4", "duration": 45.1451, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a small soccer ball is visible in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a small soccer ball is visible in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 44.0], "base_se_norm": [-0.9556984305381775, 0.9492702484130859], "gt_se_raw": [1.0, 10.0], "gt_se_norm": [-0.9556984305381775, -0.5569840669631958], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005982.pt"}
{"idx": 5984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ij9EnG6cF40.mp4", "duration": 122.41666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman in the green shirt says that Nick won't make it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman in the green shirt says that Nick won't make it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 44.0], "base_se_norm": [-0.558883547782898, -0.2811436057090759], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.558883547782898, -0.49353301525115967], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005984.pt"}
{"idx": 5986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ij9EnG6cF40.mp4", "duration": 122.41666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket explain that he is autistic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket explain that he is autistic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 68.0], "base_se_norm": [-0.1014295220375061, 0.11095988750457764], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.012933969497680664, 0.045609235763549805], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005986.pt"}
{"idx": 5988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8waFw6Y9fFM.mp4", "duration": 205.93906666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man eating icing directly from the bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man eating icing directly from the bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.7572097778320312, -0.6989400386810303], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.7572097778320312, -0.7183632850646973], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005988.pt"}
{"idx": 5990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8waFw6Y9fFM.mp4", "duration": 205.93906666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the gingerbread house collapse for the first time while he's trying to add the roof?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the gingerbread house collapse for the first time while he's trying to add the roof?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 128.0], "base_se_norm": [0.15568161010742188, 0.2430860996246338], "gt_se_raw": [114.0, 116.0], "gt_se_norm": [0.10712361335754395, 0.1265467405319214], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005990.pt"}
{"idx": 5992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8waFw6Y9fFM.mp4", "duration": 205.93906666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man get angry and smash the gingerbread house with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get angry and smash the gingerbread house with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 130.0], "base_se_norm": [0.2430860996246338, 0.2625093460083008], "gt_se_raw": [185.0, 187.0], "gt_se_norm": [0.7966479063034058, 0.8160711526870728], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005992.pt"}
{"idx": 5994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMfVl38EwpI.mp4", "duration": 110.27683333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the construction workers using a crane to install the large acrylic panels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the construction workers using a crane to install the large acrylic panels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 27.0], "base_se_norm": [-0.6372765302658081, -0.5103232860565186], "gt_se_raw": [20.0, 27.0], "gt_se_norm": [-0.6372765302658081, -0.5103232860565186], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005994.pt"}
{"idx": 5996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMfVl38EwpI.mp4", "duration": 110.27683333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man holds the acrylic cube in front of his face to show its clarity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man holds the acrylic cube in front of his face to show its clarity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.07003438472747803, 0.1425790786743164], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [0.05189824104309082, 0.1244429349899292], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005996.pt"}
{"idx": 5998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uKPD6tPry6g.mp4", "duration": 76.60986666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the black and white footage of a man speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the black and white footage of a man speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8694685101509094, -0.7389370203018188], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.8955748081207275, -0.765043318271637], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005998.pt"}
{"idx": 6000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uKPD6tPry6g.mp4", "duration": 76.60986666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are hands shown operating a sound mixing board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are hands shown operating a sound mixing board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.5300866365432739, -0.39955514669418335], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.5561928749084473, -0.42566144466400146], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006000.pt"}
{"idx": 6002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uKPD6tPry6g.mp4", "duration": 76.60986666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the close-up shot of a hand adjusting a synthesizer's controls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the close-up shot of a hand adjusting a synthesizer's controls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [0.461952805519104, 0.5402716398239136], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [0.461952805519104, 0.5141654014587402], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006002.pt"}
{"idx": 6004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aB_s9lw9E-M.mp4", "duration": 194.6945, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in black robes is delivering a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in black robes is delivering a speech.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 159.0], "base_se_norm": [-0.45555728673934937, 0.6333280801773071], "gt_se_raw": [32.0, 128.0], "gt_se_norm": [-0.6712799072265625, 0.31488049030303955], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006004.pt"}
{"idx": 6006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aB_s9lw9E-M.mp4", "duration": 194.6945, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range of the magic duel between the two characters in black.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range of the magic duel between the two characters in black.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 183.0], "base_se_norm": [0.7360531091690063, 0.8798681497573853], "gt_se_raw": [169.0, 184.0], "gt_se_norm": [0.7360531091690063, 0.8901406526565552], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006006.pt"}
{"idx": 6008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkUhEpx7E2k.mp4", "duration": 131.42295555555555, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the potatoes are being sliced with a mandoline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the potatoes are being sliced with a mandoline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 13.0], "base_se_norm": [-0.9239097833633423, -0.8021654486656189], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.9239097833633423, -0.8478195667266846], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006008.pt"}
{"idx": 6010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkUhEpx7E2k.mp4", "duration": 131.42295555555555, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the sauce is being prepared by whisking cream into the dry ingredients.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the sauce is being prepared by whisking cream into the dry ingredients.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 61.0], "base_se_norm": [-0.17822575569152832, -0.07169950008392334], "gt_se_raw": [56.0, 61.0], "gt_se_norm": [-0.14778971672058105, -0.07169950008392334], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006010.pt"}
{"idx": 6012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkUhEpx7E2k.mp4", "duration": 131.42295555555555, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a slice of the finished dish is being lifted out of the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a slice of the finished dish is being lifted out of the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 101.0], "base_se_norm": [0.4761500358581543, 0.5370221138000488], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [0.49136805534362793, 0.5370221138000488], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006012.pt"}
{"idx": 6014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kPm0_jB5EQI.mp4", "duration": 77.11036666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the family members shouting at each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the family members shouting at each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.7406314611434937, -0.6628209352493286], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.7665683627128601, -0.6368840932846069], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006014.pt"}
{"idx": 6016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kPm0_jB5EQI.mp4", "duration": 77.11036666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the family is being loaded into the van.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the family is being loaded into the van.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 25.0], "base_se_norm": [-0.5331367254257202, -0.3515787720680237], "gt_se_raw": [18.0, 25.0], "gt_se_norm": [-0.5331367254257202, -0.3515787720680237], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006016.pt"}
{"idx": 6018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Lt8XAfU_vlM.mp4", "duration": 26.026, "prompt_text": "Please find the visual event described by the sentence 'When does the boy stretch with his hands behind his neck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy stretch with his hands behind his neck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.6157687902450562], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.6926150321960449], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006018.pt"}
{"idx": 6020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Lt8XAfU_vlM.mp4", "duration": 26.026, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy take the red drawing board out of the blue basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy take the red drawing board out of the blue basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.23153764009475708, -0.0009989738464355469], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.23153764009475708, -0.0009989738464355469], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006020.pt"}
{"idx": 6022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Lt8XAfU_vlM.mp4", "duration": 26.026, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman guides the boy's left hand towards the pink basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman guides the boy's left hand towards the pink basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.46007847785949707, 0.6906170845031738], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.5369247198104858, 0.6906170845031738], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006022.pt"}
{"idx": 6024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QEpCsMbMx7w.mp4", "duration": 179.34583333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man in a beige trench coat enters the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man in a beige trench coat enters the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 74.0], "base_se_norm": [-0.2862951159477234, -0.1747787594795227], "gt_se_raw": [64.0, 73.0], "gt_se_norm": [-0.2862951159477234, -0.18593037128448486], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006024.pt"}
{"idx": 6026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QEpCsMbMx7w.mp4", "duration": 179.34583333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does a man sit down on the red bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does a man sit down on the red bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 86.0], "base_se_norm": [-0.07441401481628418, -0.040959060192108154], "gt_se_raw": [84.0, 86.0], "gt_se_norm": [-0.06326234340667725, -0.040959060192108154], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006026.pt"}
{"idx": 6028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i4eADcCnFjo.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy use the device to state that Emily is at school?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy use the device to state that Emily is at school?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [0.45469534397125244, 0.6861240863800049], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.504081130027771, -0.3387748599052429], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006028.pt"}
{"idx": 6030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i4eADcCnFjo.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher placing a paper cutout of a boy onto the playground picture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher placing a paper cutout of a boy onto the playground picture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9008162021636963, -0.7685711979866028], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.8677549362182617, -0.7355099320411682], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006030.pt"}
{"idx": 6032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i4eADcCnFjo.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy use the tablet to answer that Ethan is in the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy use the tablet to answer that Ethan is in the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [0.42163407802581787, 0.5869404077529907], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [0.8844916820526123, 0.9506142139434814], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006032.pt"}
{"idx": 6034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ccgW5CHFg0Y.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man lights a cigarette with a lighter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man lights a cigarette with a lighter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.3339993357658386, -0.20079922676086426], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.3339993357658386, -0.26739931106567383], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006034.pt"}
{"idx": 6036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ccgW5CHFg0Y.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is a drink being shaken in a cocktail shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a drink being shaken in a cocktail shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.5318014621734619, 0.665001630783081], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.7335997819900513, -0.6669996976852417], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006036.pt"}
{"idx": 6038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vRnhEjP3R-c.mp4", "duration": 162.16199999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the two women standing in the doorway of a messy bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the two women standing in the doorway of a messy bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 43.0], "base_se_norm": [-0.49433284997940063, -0.46966612339019775], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [-0.4819995164871216, -0.4573327898979187], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006038.pt"}
{"idx": 6040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vRnhEjP3R-c.mp4", "duration": 162.16199999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a woman is trying on a leopard print top, but it's on backwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a woman is trying on a leopard print top, but it's on backwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 77.0], "base_se_norm": [-0.09966576099395752, -0.05033242702484131], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [-0.09966576099395752, -0.03799903392791748], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006040.pt"}
{"idx": 6042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vRnhEjP3R-c.mp4", "duration": 162.16199999999998, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time a man in a plaid shirt falls onto a music box in a bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time a man in a plaid shirt falls onto a music box in a bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 133.0], "base_se_norm": [0.6156682968139648, 0.640334963798523], "gt_se_raw": [131.0, 140.0], "gt_se_norm": [0.6156682968139648, 0.7266683578491211], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006042.pt"}
{"idx": 6044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dsf_z4urc4s.mp4", "duration": 130.12, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy opens a drawer and takes out chalks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy opens a drawer and takes out chalks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 37.0], "base_se_norm": [-0.5235167741775513, -0.43129414319992065], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.5081462860107422, -0.43129414319992065], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006044.pt"}
{"idx": 6046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dsf_z4urc4s.mp4", "duration": 130.12, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy takes off his sandals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy takes off his sandals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 67.0], "base_se_norm": [-0.07777434587478638, 0.02981865406036377], "gt_se_raw": [60.0, 64.0], "gt_se_norm": [-0.07777434587478638, -0.01629263162612915], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006046.pt"}
{"idx": 6048, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pr4CRc_h-9w.mp4", "duration": 190.19, "prompt_text": "Please find the visual event described by the sentence 'When does the man turn off the alarm and wake up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man turn off the alarm and wake up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 18.0], "base_se_norm": [-0.9053577780723572, -0.8107156157493591], "gt_se_raw": [10.0, 18.0], "gt_se_norm": [-0.8948420286178589, -0.8107156157493591], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006048.pt"}
{"idx": 6050, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pr4CRc_h-9w.mp4", "duration": 190.19, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the mother takes a photo of her family.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the mother takes a photo of her family.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [-0.3900836110115051, -0.33750462532043457], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.3900836110115051, -0.3585361838340759], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006050.pt"}
{"idx": 6052, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pr4CRc_h-9w.mp4", "duration": 190.19, "prompt_text": "Please find the visual event described by the sentence 'Show the father lying in bed, looking at the photos his daughter has put up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the father lying in bed, looking at the photos his daughter has put up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 168.0], "base_se_norm": [0.5247910022735596, 0.766654372215271], "gt_se_raw": [146.0, 164.0], "gt_se_norm": [0.5353068113327026, 0.7245911359786987], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006052.pt"}
{"idx": 6054, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CwwZOzZtFAM.mp4", "duration": 88.25483333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the driver correctly offers assistance to the woman with the service dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the driver correctly offers assistance to the woman with the service dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 35.0], "base_se_norm": [-0.45612049102783203, -0.2068423628807068], "gt_se_raw": [25.0, 34.0], "gt_se_norm": [-0.4334588050842285, -0.22950398921966553], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006054.pt"}
{"idx": 6056, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CwwZOzZtFAM.mp4", "duration": 88.25483333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the woman get upset and complain that this is the fourth time a driver has left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman get upset and complain that this is the fourth time a driver has left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [-0.02554917335510254, 0.06509745121002197], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.02554917335510254, 0.06509745121002197], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006056.pt"}
{"idx": 6058, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7Fjt-mlIlTY.mp4", "duration": 49.315933333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the adult's hand place the yellow squeaky toy into the metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the adult's hand place the yellow squeaky toy into the metal bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.7161160707473755, -0.5944515466690063], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.7161160707473755, -0.5538967251777649], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006058.pt"}
{"idx": 6060, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7Fjt-mlIlTY.mp4", "duration": 49.315933333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the baby start hitting the toy inside the bowl with the whisk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the baby start hitting the toy inside the bowl with the whisk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 42.0], "base_se_norm": [0.25720012187957764, 0.7033034563064575], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [0.29775500297546387, 0.45997440814971924], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006060.pt"}
{"idx": 6062, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7Fjt-mlIlTY.mp4", "duration": 49.315933333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the baby holding the whisk but looking away to her right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the baby holding the whisk but looking away to her right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.7566709518432617], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.0266837477684021, 0.09498083591461182], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006062.pt"}
{"idx": 6064, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y7HWlprfaGA.mp4", "duration": 15.666666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the character sits in front of the computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the character sits in front of the computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.4893617033958435, -0.23404258489608765], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.6170213222503662, -0.3617021441459656], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006064.pt"}
{"idx": 6066, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y7HWlprfaGA.mp4", "duration": 15.666666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the phrase 'Describe what you see' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the phrase 'Describe what you see' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [0.4042552709579468, 0.7872339487075806], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [0.4042552709579468, 0.7872339487075806], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006066.pt"}
{"idx": 6068, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IWrK7sXK3Vw.mp4", "duration": 211.6116116116116, "prompt_text": "Please find the visual event described by the sentence 'When is the teddy bear hugging the bunny?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teddy bear hugging the bunny?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 18.0], "base_se_norm": [-1.0, -0.8298770189285278], "gt_se_raw": [1.0, 18.0], "gt_se_norm": [-0.9905487298965454, -0.8298770189285278], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006068.pt"}
{"idx": 6070, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IWrK7sXK3Vw.mp4", "duration": 211.6116116116116, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the bunny interacts with its reflection in the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the bunny interacts with its reflection in the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 84.0], "base_se_norm": [-0.5841438174247742, -0.2060927152633667], "gt_se_raw": [45.0, 84.0], "gt_se_norm": [-0.5746925473213196, -0.2060927152633667], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006070.pt"}
{"idx": 6072, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IWrK7sXK3Vw.mp4", "duration": 211.6116116116116, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the bunny is playing with a pink macaron-shaped object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the bunny is playing with a pink macaron-shaped object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 201.0], "base_se_norm": [0.7484862804412842, 0.8997066020965576], "gt_se_raw": [186.0, 201.0], "gt_se_norm": [0.7579375505447388, 0.8997066020965576], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006072.pt"}
{"idx": 6074, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7a4yAbsPCQg.mp4", "duration": 84.618, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention memory, emotions, and fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention memory, emotions, and fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [-0.12548154592514038, 0.08723914623260498], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.4091091752052307, -0.31456661224365234], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006074.pt"}
{"idx": 6076, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7a4yAbsPCQg.mp4", "duration": 84.618, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker talk about developing a relationship with a deaf-blind student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker talk about developing a relationship with a deaf-blind student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.5745586156845093, -0.38547348976135254], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.03093898296356201, 0.08723914623260498], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006076.pt"}
{"idx": 6078, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hDUsLqVeyy4.mp4", "duration": 75.34193333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a black shirt gesturing with both hands while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a black shirt gesturing with both hands while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 22.0], "base_se_norm": [-0.7079979181289673, -0.41599589586257935], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.6814522743225098, -0.5487240552902222], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006078.pt"}
{"idx": 6080, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hDUsLqVeyy4.mp4", "duration": 75.34193333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman with sunglasses on her head lean in and point at the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman with sunglasses on her head lean in and point at the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [0.22109949588775635, 0.38037335872650146], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [0.24764513969421387, 0.3272820711135864], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006080.pt"}
{"idx": 6082, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hDUsLqVeyy4.mp4", "duration": 75.34193333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the boy is so overcome with laughter that he lays his head down on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the boy is so overcome with laughter that he lays his head down on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 74.0], "base_se_norm": [0.8316493034362793, 0.9643775224685669], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [0.8316493034362793, 0.9112862348556519], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006082.pt"}
{"idx": 6084, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g5Bfey0nuGI.mp4", "duration": 169.29412499999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot of the hotel's reception desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot of the hotel's reception desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.8700486421585083, -0.8346073627471924], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [-0.8700486421585083, -0.8582348823547363], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006084.pt"}
{"idx": 6086, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g5Bfey0nuGI.mp4", "duration": 169.29412499999998, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the trainers are doing plank push-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the trainers are doing plank push-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 123.0], "base_se_norm": [-0.13759559392929077, 0.4530923366546631], "gt_se_raw": [95.0, 110.0], "gt_se_norm": [0.12230706214904785, 0.29951345920562744], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006086.pt"}
{"idx": 6088, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gVt9WTJkk7g.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the exterior shot of the Laboratory Sciences Annex.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the exterior shot of the Laboratory Sciences Annex.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8653032183647156], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8653032183647156], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006088.pt"}
{"idx": 6090, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gVt9WTJkk7g.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the researcher examining a white, steaming object with tweezers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the researcher examining a white, steaming object with tweezers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [-0.15814518928527832, -0.0907968282699585], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.15814518928527832, -0.0907968282699585], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006090.pt"}
{"idx": 6092, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gVt9WTJkk7g.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a teacher is writing on a whiteboard in an empty classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a teacher is writing on a whiteboard in an empty classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [0.48166441917419434, 0.616361141204834], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.48166441917419434, 0.5826870203018188], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006092.pt"}
{"idx": 6094, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yXQb9MHDvnI.mp4", "duration": 66.291, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl successfully open the velcro strap on her shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl successfully open the velcro strap on her shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.517279863357544, -0.39659982919692993], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.637959897518158, -0.48710983991622925], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006094.pt"}
{"idx": 6096, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yXQb9MHDvnI.mp4", "duration": 66.291, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the girl is fastening the shoe strap at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the girl is fastening the shoe strap at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 65.0], "base_se_norm": [0.5990104675292969, 0.9610505104064941], "gt_se_raw": [54.0, 65.0], "gt_se_norm": [0.6291804313659668, 0.9610505104064941], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006096.pt"}
{"idx": 6098, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tmg5hLMC1s.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker state that the employment rate is higher for one group than another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker state that the employment rate is higher for one group than another?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.03166300058364868], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.6126651763916016, -0.4189978241920471], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006098.pt"}
{"idx": 6100, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tmg5hLMC1s.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the 'Non-disabled' category mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'Non-disabled' category mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.8063325881958008, -0.6126651763916016], "start_token_span": [70, 73], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006100.pt"}
{"idx": 6102, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tmg5hLMC1s.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narration begin to explain the chart titled 'Employment for disabled and non-disabled Americans'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narration begin to explain the chart titled 'Employment for disabled and non-disabled Americans'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.03166300058364868], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.8063325881958008], "start_token_span": [83, 86], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006102.pt"}
{"idx": 6104, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IKs5-pkF4qE.mp4", "duration": 141.52, "prompt_text": "Please find the visual event described by the sentence 'Show the part where an example of a person with Parkinson's disease is given.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where an example of a person with Parkinson's disease is given.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 46.0], "base_se_norm": [-0.5195025205612183, -0.34991520643234253], "gt_se_raw": [33.0, 46.0], "gt_se_norm": [-0.5336348414421082, -0.34991520643234253], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006104.pt"}
{"idx": 6106, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IKs5-pkF4qE.mp4", "duration": 141.52, "prompt_text": "Please find the visual event described by the sentence 'Show the demonstration of the 'Lucy 4' adaptive keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the demonstration of the 'Lucy 4' adaptive keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 103.0], "base_se_norm": [0.30016958713531494, 0.4556245803833008], "gt_se_raw": [92.0, 102.0], "gt_se_norm": [0.30016958713531494, 0.44149231910705566], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006106.pt"}
{"idx": 6108, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7WhJ2L5xUqI.mp4", "duration": 54.58786666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the blue striped shirt speaking at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the blue striped shirt speaking at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.7801709175109863], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.7801709175109863], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006108.pt"}
{"idx": 6110, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7WhJ2L5xUqI.mp4", "duration": 54.58786666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 3D print of 'Washington Crossing the Delaware' with embedded audio sensors shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 3D print of 'Washington Crossing the Delaware' with embedded audio sensors shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 21.0], "base_se_norm": [-0.41378915309906006, -0.23059827089309692], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [-0.41378915309906006, -0.23059827089309692], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006110.pt"}
{"idx": 6112, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7WhJ2L5xUqI.mp4", "duration": 54.58786666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the '3DPHOTOWORKS' company logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the '3DPHOTOWORKS' company logo appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [0.8685470819473267, 0.9784616231918335], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [0.8685470819473267, 0.9784616231918335], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006112.pt"}
{"idx": 6114, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zaSqDJMX_nc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the child puts their hand in the bowl to grab food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the child puts their hand in the bowl to grab food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.8688400983810425, -0.606520414352417], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [0.11485886573791504, 0.3771786689758301], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006114.pt"}
{"idx": 6116, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zaSqDJMX_nc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman is feeding the child with the red spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman is feeding the child with the red spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [0.5739184617996216, 0.9673980474472046], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.606520414352417, -0.4753605127334595], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006116.pt"}
{"idx": 6118, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tKB4h9gvmm0.mp4", "duration": 112.0, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a character tricks another into looking up at the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a character tricks another into looking up at the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.7142857313156128, -0.6428571343421936], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.7142857313156128, -0.6428571343421936], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006118.pt"}
{"idx": 6120, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tKB4h9gvmm0.mp4", "duration": 112.0, "prompt_text": "Please find the visual event described by the sentence 'When does a llama drive a car off a cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a llama drive a car off a cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 35.0], "base_se_norm": [-0.4464285969734192, -0.375], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.4464285969734192, -0.375], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006120.pt"}
{"idx": 6122, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tKB4h9gvmm0.mp4", "duration": 112.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man dies from an allergy to a cute kitten.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man dies from an allergy to a cute kitten.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 77.0], "base_se_norm": [0.3035714626312256, 0.375], "gt_se_raw": [70.0, 77.0], "gt_se_norm": [0.25, 0.375], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006122.pt"}
{"idx": 6124, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R3tpOXNNMTc.mp4", "duration": 279.8212083333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man wearing a red shirt is playing table tennis.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 217.0 - 223.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man wearing a red shirt is playing table tennis.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 217.0 - 223.0 seconds.<|im_end|>\n", "base_se_raw": [217.0, 223.0], "base_se_norm": [0.5509904623031616, 0.5938749313354492], "gt_se_raw": [217.0, 222.0], "gt_se_norm": [0.5509904623031616, 0.586727499961853], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006124.pt"}
{"idx": 6126, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R3tpOXNNMTc.mp4", "duration": 279.8212083333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man feeding ducks by the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 232.0 - 241.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man feeding ducks by the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 232.0 - 241.0 seconds.<|im_end|>\n", "base_se_raw": [232.0, 241.0], "base_se_norm": [0.6582016944885254, 0.7225285768508911], "gt_se_raw": [232.0, 241.0], "gt_se_norm": [0.6582016944885254, 0.7225285768508911], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006126.pt"}
{"idx": 6128, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PJlmYh27MHg.mp4", "duration": 208.7085, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the blue shirt take off his hat and sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the blue shirt take off his hat and sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 17.0], "base_se_norm": [-0.9137552976608276, -0.8370933532714844], "gt_se_raw": [8.0, 18.0], "gt_se_norm": [-0.9233380556106567, -0.8275105953216553], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006128.pt"}
{"idx": 6130, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PJlmYh27MHg.mp4", "duration": 208.7085, "prompt_text": "Please find the visual event described by the sentence 'When can we see the first full shot of the dinosaur's body?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the first full shot of the dinosaur's body?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 41.0], "base_se_norm": [-0.6837694644927979, -0.6071075201034546], "gt_se_raw": [33.0, 41.0], "gt_se_norm": [-0.6837694644927979, -0.6071075201034546], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006130.pt"}
{"idx": 6132, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PJlmYh27MHg.mp4", "duration": 208.7085, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the dinosaur stands on its hind legs to eat from a tall tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the dinosaur stands on its hind legs to eat from a tall tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 96.0], "base_se_norm": [-0.16630131006240845, -0.08005660772323608], "gt_se_raw": [87.0, 97.0], "gt_se_norm": [-0.16630131006240845, -0.07047384977340698], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006132.pt"}
{"idx": 6134, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQHxqp2H0qI.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man uses the watch to cheat on a test.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man uses the watch to cheat on a test.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 82.0], "base_se_norm": [-0.6200000047683716, -0.45333331823349], "gt_se_raw": [59.0, 81.0], "gt_se_norm": [-0.6066666841506958, -0.4599999785423279], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006134.pt"}
{"idx": 6136, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQHxqp2H0qI.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the pink shirt drags the other man into the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 206.0 - 211.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the pink shirt drags the other man into the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 206.0 - 211.0 seconds.<|im_end|>\n", "base_se_raw": [206.0, 211.0], "base_se_norm": [0.3733333349227905, 0.40666663646698], "gt_se_raw": [202.0, 211.0], "gt_se_norm": [0.34666669368743896, 0.40666663646698], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006136.pt"}
{"idx": 6138, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nWjEYlMZnhE.mp4", "duration": 238.4, "prompt_text": "Please find the visual event described by the sentence 'Show the process of folding the paper into a triangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of folding the paper into a triangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 52.0], "base_se_norm": [-0.781879186630249, -0.563758373260498], "gt_se_raw": [26.0, 48.0], "gt_se_norm": [-0.781879186630249, -0.5973154306411743], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006138.pt"}
{"idx": 6140, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nWjEYlMZnhE.mp4", "duration": 238.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does the paper get collapsed into a smaller, double-layered triangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the paper get collapsed into a smaller, double-layered triangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 102.0], "base_se_norm": [-0.2533556818962097, -0.14429527521133423], "gt_se_raw": [89.0, 101.0], "gt_se_norm": [-0.2533556818962097, -0.15268456935882568], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006140.pt"}
{"idx": 6142, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nWjEYlMZnhE.mp4", "duration": 238.4, "prompt_text": "Please find the visual event described by the sentence 'When are the wings of the butterfly folded out to create the final shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 180.0 - 218.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the wings of the butterfly folded out to create the final shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 180.0 - 218.0 seconds.<|im_end|>\n", "base_se_raw": [180.0, 218.0], "base_se_norm": [0.5100671052932739, 0.8288590908050537], "gt_se_raw": [175.0, 187.0], "gt_se_norm": [0.468120813369751, 0.5687919855117798], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006142.pt"}
{"idx": 6144, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-AKDXFx4jI.mp4", "duration": 201.33446666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a woman with a pinstripe jacket and a butterfly necklace is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a woman with a pinstripe jacket and a butterfly necklace is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 61.0], "base_se_norm": [-0.5430489778518677, -0.3940431475639343], "gt_se_raw": [46.0, 61.0], "gt_se_norm": [-0.5430489778518677, -0.3940431475639343], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006144.pt"}
{"idx": 6146, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-AKDXFx4jI.mp4", "duration": 201.33446666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a close-up of a hand touching the decorations on the piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 146.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a close-up of a hand touching the decorations on the piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 146.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [146.0, 149.0], "base_se_norm": [0.45032286643981934, 0.4801241159439087], "gt_se_raw": [146.0, 149.0], "gt_se_norm": [0.45032286643981934, 0.4801241159439087], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006146.pt"}
{"idx": 6148, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1cHSTAv6aw.mp4", "duration": 227.39383333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the first instance of the boy's hands typing on the laptop keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the first instance of the boy's hands typing on the laptop keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.6921640634536743, -0.6657781004905701], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.700959324836731, -0.6745734214782715], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006148.pt"}
{"idx": 6150, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1cHSTAv6aw.mp4", "duration": 227.39383333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is the word 'example' completed using the word prediction software?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the word 'example' completed using the word prediction software?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 86.0], "base_se_norm": [-0.269989013671875, -0.24360305070877075], "gt_se_raw": [83.0, 86.0], "gt_se_norm": [-0.269989013671875, -0.24360305070877075], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006150.pt"}
{"idx": 6152, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1cHSTAv6aw.mp4", "duration": 227.39383333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the boy getting up from his chair and walking away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 202.0 - 206.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the boy getting up from his chair and walking away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 202.0 - 206.0 seconds.<|im_end|>\n", "base_se_raw": [202.0, 206.0], "base_se_norm": [0.7766532897949219, 0.8118345737457275], "gt_se_raw": [203.0, 206.0], "gt_se_norm": [0.7854486703872681, 0.8118345737457275], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006152.pt"}
{"idx": 6154, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBF5l9rWQFs.mp4", "duration": 286.625, "prompt_text": "Please find the visual event described by the sentence 'Show the animation of a woman in a purple dress dancing with a man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the animation of a woman in a purple dress dancing with a man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.8255560398101807, -0.7906672358512878], "gt_se_raw": [25.0, 30.0], "gt_se_norm": [-0.8255560398101807, -0.7906672358512878], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006154.pt"}
{"idx": 6156, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBF5l9rWQFs.mp4", "duration": 286.625, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman with voluminous black hair is speaking passionately.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman with voluminous black hair is speaking passionately.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 112.0], "base_se_norm": [-0.27431315183639526, -0.2184910774230957], "gt_se_raw": [120.0, 125.0], "gt_se_norm": [-0.16266900300979614, -0.12778019905090332], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006156.pt"}
{"idx": 6158, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mwm0OwqWvF4.mp4", "duration": 305.56, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman helping a boy get on the dog's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman helping a boy get on the dog's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [-0.8101845979690552, -0.7709124088287354], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.7970938682556152, -0.7709124088287354], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006158.pt"}
{"idx": 6160, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mwm0OwqWvF4.mp4", "duration": 305.56, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog start eating the meat from the silver bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 258.0 - 263.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog start eating the meat from the silver bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 258.0 - 263.0 seconds.<|im_end|>\n", "base_se_raw": [258.0, 263.0], "base_se_norm": [0.6887027025222778, 0.7214294672012329], "gt_se_raw": [258.0, 262.0], "gt_se_norm": [0.6887027025222778, 0.7148841619491577], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006160.pt"}
{"idx": 6162, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mwm0OwqWvF4.mp4", "duration": 305.56, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the dog is shown climbing a wooden wall during training.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the dog is shown climbing a wooden wall during training.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [-0.6203691959381104, -0.5941877365112305], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.6138237714767456, -0.5941877365112305], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006162.pt"}
{"idx": 6164, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ygqARUR7aW8.mp4", "duration": 228.0278, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman mention 'World Parrot Day'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman mention 'World Parrot Day'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 18.0], "base_se_norm": [-0.9824582934379578, -0.8421245217323303], "gt_se_raw": [2.0, 8.0], "gt_se_norm": [-0.9824582934379578, -0.9298331141471863], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006164.pt"}
{"idx": 6166, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ygqARUR7aW8.mp4", "duration": 228.0278, "prompt_text": "Please find the visual event described by the sentence 'When does the woman talk about 'Zygodactyl' feet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman talk about 'Zygodactyl' feet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 41.0], "base_se_norm": [-0.7456450462341309, -0.6403946876525879], "gt_se_raw": [29.0, 38.0], "gt_se_norm": [-0.7456450462341309, -0.6667072772979736], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006166.pt"}
{"idx": 6168, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ijCJg0-nL-U.mp4", "duration": 297.297, "prompt_text": "Please find the visual event described by the sentence 'When is the KISS GEL EYELINER shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the KISS GEL EYELINER shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [-0.811636209487915, -0.7914543151855469], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [-0.811636209487915, -0.7914543151855469], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006168.pt"}
{"idx": 6170, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ijCJg0-nL-U.mp4", "duration": 297.297, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the L'OREAL TELESCOPIC MASCARA is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the L'OREAL TELESCOPIC MASCARA is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 128.0], "base_se_norm": [-0.1590900421142578, -0.13890820741653442], "gt_se_raw": [125.0, 128.0], "gt_se_norm": [-0.1590900421142578, -0.13890820741653442], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006170.pt"}
{"idx": 6172, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ijCJg0-nL-U.mp4", "duration": 297.297, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman is using a makeup sponge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 189.0 - 207.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman is using a makeup sponge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 189.0 - 207.0 seconds.<|im_end|>\n", "base_se_raw": [189.0, 207.0], "base_se_norm": [0.27145588397979736, 0.3925468921661377], "gt_se_raw": [190.0, 200.0], "gt_se_norm": [0.27818310260772705, 0.3454558849334717], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006172.pt"}
{"idx": 6174, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TuE6z8X-rGg.mp4", "duration": 303.7034, "prompt_text": "Please find the visual event described by the sentence 'A woman with blonde hair teaches another person by guiding her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman with blonde hair teaches another person by guiding her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 108.0], "base_se_norm": [-0.42707258462905884, -0.2887797951698303], "gt_se_raw": [88.0, 108.0], "gt_se_norm": [-0.4204872250556946, -0.2887797951698303], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006174.pt"}
{"idx": 6176, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TuE6z8X-rGg.mp4", "duration": 303.7034, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment where a little girl is crawling through an orange tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 210.0 - 220.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment where a little girl is crawling through an orange tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 210.0 - 220.0 seconds.<|im_end|>\n", "base_se_raw": [210.0, 220.0], "base_se_norm": [0.3829282522201538, 0.44878196716308594], "gt_se_raw": [210.0, 219.0], "gt_se_norm": [0.3829282522201538, 0.4421966075897217], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006176.pt"}
{"idx": 6178, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vN1fuDgj-Qk.mp4", "duration": 222.18029166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman present the earrings she is wearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman present the earrings she is wearing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 65.0], "base_se_norm": [-0.47790151834487915, -0.4148896336555481], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.46889978647232056, -0.4148896336555481], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006178.pt"}
{"idx": 6180, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vN1fuDgj-Qk.mp4", "duration": 222.18029166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman is displaying the inner lining of her plaid jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman is displaying the inner lining of her plaid jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 161.0], "base_se_norm": [0.3952631950378418, 0.4492734670639038], "gt_se_raw": [158.0, 161.0], "gt_se_norm": [0.422268271446228, 0.4492734670639038], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006180.pt"}
{"idx": 6182, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vN1fuDgj-Qk.mp4", "duration": 222.18029166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where a woman wearing a black coat and a beige sweater poses for the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where a woman wearing a black coat and a beige sweater poses for the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.9099830389022827, -0.8829779028892517], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.9099830389022827, -0.8829779028892517], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006182.pt"}
{"idx": 6184, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ekpylglrh2U.mp4", "duration": 313.7134, "prompt_text": "Please find the visual event described by the sentence 'When does the man with the mustache raise his glass for a toast?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man with the mustache raise his glass for a toast?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [-0.47085464000701904, -0.45810413360595703], "gt_se_raw": [87.0, 89.0], "gt_se_norm": [-0.445353627204895, -0.4326031804084778], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006184.pt"}
{"idx": 6186, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ekpylglrh2U.mp4", "duration": 313.7134, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman in the red robe is giving a speech on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 204.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman in the red robe is giving a speech on stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 204.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 204.0], "base_se_norm": [0.02003931999206543, 0.30055010318756104], "gt_se_raw": [160.0, 165.0], "gt_se_norm": [0.02003931999206543, 0.051915526390075684], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006186.pt"}
{"idx": 6188, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QVwXvsrJ6UA.mp4", "duration": 274.73333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a group of young friends are gathered on a wooden dock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a group of young friends are gathered on a wooden dock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [-0.6432904601097107, -0.5923320055007935], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [-0.6432904601097107, -0.599611759185791], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006188.pt"}
{"idx": 6190, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QVwXvsrJ6UA.mp4", "duration": 274.73333333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see three graduates holding their diplomas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see three graduates holding their diplomas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 121.0], "base_se_norm": [-0.1628245711326599, -0.11914587020874023], "gt_se_raw": [114.0, 122.0], "gt_se_norm": [-0.17010438442230225, -0.1118660569190979], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006190.pt"}
{"idx": 6192, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QVwXvsrJ6UA.mp4", "duration": 274.73333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the picture where a mother is pointing her finger at a boy on a couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the picture where a mother is pointing her finger at a boy on a couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 96.0], "base_se_norm": [-0.3520990014076233, -0.3011404871940613], "gt_se_raw": [89.0, 96.0], "gt_se_norm": [-0.3520990014076233, -0.3011404871940613], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006192.pt"}
{"idx": 6194, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mO3mMYwKkKs.mp4", "duration": 325.24, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the video introduces the tools known as 'Lock Picks'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the video introduces the tools known as 'Lock Picks'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 162.0], "base_se_norm": [-0.13909727334976196, -0.003812551498413086], "gt_se_raw": [140.0, 148.0], "gt_se_norm": [-0.13909727334976196, -0.08990281820297241], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006194.pt"}
{"idx": 6196, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mO3mMYwKkKs.mp4", "duration": 325.24, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the 'single pin picking' method is used to successfully open the transparent lock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 213.0 - 238.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the 'single pin picking' method is used to successfully open the transparent lock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 213.0 - 238.0 seconds.<|im_end|>\n", "base_se_raw": [213.0, 238.0], "base_se_norm": [0.3098020553588867, 0.46353471279144287], "gt_se_raw": [233.0, 239.0], "gt_se_norm": [0.4327881336212158, 0.46968400478363037], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006196.pt"}
{"idx": 6198, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mvD5wvlIE7c.mp4", "duration": 230.8973333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the view from the airplane window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the view from the airplane window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [-0.8181009888648987, -0.8007772564888], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [-0.8094391226768494, -0.8007772564888], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006198.pt"}
{"idx": 6200, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mvD5wvlIE7c.mp4", "duration": 230.8973333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the students seen working diligently in their hotel room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the students seen working diligently in their hotel room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 69.0], "base_se_norm": [-0.5149359703063965, -0.40233176946640015], "gt_se_raw": [61.0, 66.0], "gt_se_norm": [-0.4716266393661499, -0.4283173680305481], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006200.pt"}
{"idx": 6202, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mvD5wvlIE7c.mp4", "duration": 230.8973333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the team jump up and celebrate their second-place win?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 196.0 - 205.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the team jump up and celebrate their second-place win?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 196.0 - 205.0 seconds.<|im_end|>\n", "base_se_raw": [196.0, 205.0], "base_se_norm": [0.6977242231369019, 0.7756808996200562], "gt_se_raw": [195.0, 199.0], "gt_se_norm": [0.6890623569488525, 0.7237098217010498], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006202.pt"}
{"idx": 6204, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ezg4sr67OGA.mp4", "duration": 257.67408333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in the beret pointing a gun at the other man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in the beret pointing a gun at the other man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 164.0], "base_se_norm": [0.17202317714691162, 0.27292585372924805], "gt_se_raw": [150.0, 153.0], "gt_se_norm": [0.16426146030426025, 0.1875467300415039], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006204.pt"}
{"idx": 6206, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ezg4sr67OGA.mp4", "duration": 257.67408333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man with the long hair catches a bullet with his bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man with the long hair catches a bullet with his bare hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 182.0], "base_se_norm": [0.38935208320617676, 0.41263723373413086], "gt_se_raw": [207.0, 209.0], "gt_se_norm": [0.6066808700561523, 0.6222043037414551], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006206.pt"}
{"idx": 6208, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7N98N0GkGjY.mp4", "duration": 270.10316666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when children are riding on a carousel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when children are riding on a carousel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.8000763654708862, -0.7630535364151001], "gt_se_raw": [23.0, 36.0], "gt_se_norm": [-0.8296947479248047, -0.7334352135658264], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006208.pt"}
{"idx": 6210, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7N98N0GkGjY.mp4", "duration": 270.10316666666665, "prompt_text": "Please find the visual event described by the sentence 'When does a man in black appear on the balcony?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man in black appear on the balcony?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [-0.4446566700935364, -0.4224429130554199], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [-0.4446566700935364, -0.4224429130554199], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006210.pt"}
{"idx": 6212, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7N98N0GkGjY.mp4", "duration": 270.10316666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a man is writing a note under a tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 227.0 - 242.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a man is writing a note under a tree.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 227.0 - 242.0 seconds.<|im_end|>\n", "base_se_raw": [227.0, 242.0], "base_se_norm": [0.6808391809463501, 0.791907787322998], "gt_se_raw": [235.0, 242.0], "gt_se_norm": [0.7400758266448975, 0.791907787322998], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006212.pt"}
{"idx": 6214, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YXOapWsVmFw.mp4", "duration": 268.47654166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a woman is seen working with wires on a breadboard at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a woman is seen working with wires on a breadboard at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.8584606051445007, -0.8063145279884338], "gt_se_raw": [72.0, 76.0], "gt_se_norm": [-0.4636402726173401, -0.4338424801826477], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006214.pt"}
{"idx": 6216, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YXOapWsVmFw.mp4", "duration": 268.47654166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the musician mention her plans to teach elementary music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the musician mention her plans to teach elementary music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 47.0], "base_se_norm": [-0.7392696142196655, -0.6498762965202332], "gt_se_raw": [213.0, 217.0], "gt_se_norm": [0.5867308378219604, 0.6165286302566528], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006216.pt"}
{"idx": 6218, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a4hqZD5g_yQ.mp4", "duration": 231.866, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a group of men walking together indoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a group of men walking together indoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 23.0], "base_se_norm": [-0.9309946298599243, -0.8016095161437988], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.939620316028595, -0.913743257522583], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006218.pt"}
{"idx": 6220, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a4hqZD5g_yQ.mp4", "duration": 231.866, "prompt_text": "Please find the visual event described by the sentence 'Show me the man wearing a t-shirt with the British flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man wearing a t-shirt with the British flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 119.0], "base_se_norm": [-0.13743281364440918, 0.026454925537109375], "gt_se_raw": [100.0, 119.0], "gt_se_norm": [-0.13743281364440918, 0.026454925537109375], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006220.pt"}
{"idx": 6222, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a4hqZD5g_yQ.mp4", "duration": 231.866, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a black t-shirt points at a man wearing a hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a black t-shirt points at a man wearing a hat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 112.0], "base_se_norm": [-0.05980175733566284, -0.033924758434295654], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [-0.05117607116699219, -0.033924758434295654], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006222.pt"}
{"idx": 6224, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l1YSzo9n8EE.mp4", "duration": 273.6066666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman spreading butter on the bread?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman spreading butter on the bread?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 98.0], "base_se_norm": [-0.6345118284225464, -0.2836431860923767], "gt_se_raw": [51.0, 102.0], "gt_se_norm": [-0.627202033996582, -0.25440412759780884], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006224.pt"}
{"idx": 6226, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l1YSzo9n8EE.mp4", "duration": 273.6066666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman cut the finished grilled cheese sandwich in half?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 221.0 - 228.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman cut the finished grilled cheese sandwich in half?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 221.0 - 228.0 seconds.<|im_end|>\n", "base_se_raw": [221.0, 228.0], "base_se_norm": [0.6154577732086182, 0.6666260957717896], "gt_se_raw": [220.0, 227.0], "gt_se_norm": [0.6081479787826538, 0.6593163013458252], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006226.pt"}
{"idx": 6228, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVMAwGXe3BY.mp4", "duration": 296.1959, "prompt_text": "Please find the visual event described by the sentence 'At what point does the second pilot make his entrance into the cockpit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the second pilot make his entrance into the cockpit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 67.0], "base_se_norm": [-0.7704222202301025, -0.5475966930389404], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [-0.7636699080467224, -0.723156213760376], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006228.pt"}
{"idx": 6230, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVMAwGXe3BY.mp4", "duration": 296.1959, "prompt_text": "Please find the visual event described by the sentence 'When does the plane lift off from the runway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 170.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the plane lift off from the runway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 170.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 170.0], "base_se_norm": [0.12087976932525635, 0.14788901805877686], "gt_se_raw": [165.0, 171.0], "gt_se_norm": [0.114127516746521, 0.1546412706375122], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006230.pt"}
{"idx": 6232, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVMAwGXe3BY.mp4", "duration": 296.1959, "prompt_text": "Please find the visual event described by the sentence 'When can we hear the alarm sound in the cockpit, just before the pilot takes control?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 166.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we hear the alarm sound in the cockpit, just before the pilot takes control?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 166.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 166.0], "base_se_norm": [0.10062301158905029, 0.12087976932525635], "gt_se_raw": [262.0, 265.0], "gt_se_norm": [0.7690994739532471, 0.7893563508987427], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006232.pt"}
{"idx": 6234, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vVdW9CGI3Nc.mp4", "duration": 221.45456666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does Ernie bring out his rubber ducky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Ernie bring out his rubber ducky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.692939281463623, -0.656814455986023], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.692939281463623, -0.656814455986023], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006234.pt"}
{"idx": 6236, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vVdW9CGI3Nc.mp4", "duration": 221.45456666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a bowling ball appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a bowling ball appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 78.0], "base_se_norm": [-0.34072256088256836, -0.2955665588378906], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [-0.3226601481437683, -0.2955665588378906], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006236.pt"}
{"idx": 6238, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZiz1i8EpSk.mp4", "duration": 212.6, "prompt_text": "Please find the visual event described by the sentence 'When can we see a damaged sign that says 'Canal Park'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a damaged sign that says 'Canal Park'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.661335825920105, -0.6237064599990845], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [-0.6707432270050049, -0.6237064599990845], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006238.pt"}
{"idx": 6240, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZiz1i8EpSk.mp4", "duration": 212.6, "prompt_text": "Please find the visual event described by the sentence 'Show the interview with the man named David Seabert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the interview with the man named David Seabert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 61.0], "base_se_norm": [-0.49200379848480225, -0.42615240812301636], "gt_se_raw": [54.0, 61.0], "gt_se_norm": [-0.49200379848480225, -0.42615240812301636], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006240.pt"}
{"idx": 6242, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZiz1i8EpSk.mp4", "duration": 212.6, "prompt_text": "Please find the visual event described by the sentence 'At what point do rescuers talk to an elderly couple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do rescuers talk to an elderly couple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 161.0], "base_se_norm": [0.3546565771102905, 0.5145813226699829], "gt_se_raw": [144.0, 162.0], "gt_se_norm": [0.3546565771102905, 0.5239887237548828], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006242.pt"}
{"idx": 6244, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RnckZWnihFQ.mp4", "duration": 303.16953333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker give an example about repairing a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker give an example about repairing a car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 145.0], "base_se_norm": [-0.14239400625228882, -0.04343944787979126], "gt_se_raw": [16.0, 35.0], "gt_se_norm": [-0.8944485187530518, -0.7691060900688171], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006244.pt"}
{"idx": 6246, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RnckZWnihFQ.mp4", "duration": 303.16953333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'How People Can Demonstrate Dependability at Work.' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'How People Can Demonstrate Dependability at Work.' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 168.0], "base_se_norm": [0.08190298080444336, 0.10829079151153564], "gt_se_raw": [164.0, 169.0], "gt_se_norm": [0.08190298080444336, 0.11488783359527588], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006246.pt"}
{"idx": 6248, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/snSwz_iPbQA.mp4", "duration": 228.10287499999998, "prompt_text": "Please find the visual event described by the sentence 'When is a man recording the sound of a fire hydrant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man recording the sound of a fire hydrant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.9386242032051086], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9912320375442505, -0.9386242032051086], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006248.pt"}
{"idx": 6250, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/snSwz_iPbQA.mp4", "duration": 228.10287499999998, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man swings his microphone like a tennis racket on a court.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 156.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man swings his microphone like a tennis racket on a court.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 156.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [156.0, 163.0], "base_se_norm": [0.3678039312362671, 0.42917966842651367], "gt_se_raw": [157.0, 162.0], "gt_se_norm": [0.3765718936920166, 0.42041170597076416], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006250.pt"}
{"idx": 6252, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/snSwz_iPbQA.mp4", "duration": 228.10287499999998, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a tape recorder with 'Labor of Love' written on it is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a tape recorder with 'Labor of Love' written on it is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.7895686626434326, -0.7632647156715393], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7895686626434326, -0.7632647156715393], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006252.pt"}
{"idx": 6254, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3dHp0FdWMI0.mp4", "duration": 262.9, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the girl is exploring the ball with her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the girl is exploring the ball with her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 62.0], "base_se_norm": [-0.7945986986160278, -0.5283377170562744], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.581589937210083, -0.558767557144165], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006254.pt"}
{"idx": 6256, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3dHp0FdWMI0.mp4", "duration": 262.9, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the girl is holding the mobility cane to her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 184.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the girl is holding the mobility cane to her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 184.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 184.0], "base_se_norm": [0.20958542823791504, 0.3997718095779419], "gt_se_raw": [159.0, 163.0], "gt_se_norm": [0.20958542823791504, 0.24001526832580566], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006256.pt"}
{"idx": 6258, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8O5vcM_00t0.mp4", "duration": 231.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the person stir the mixture with a straw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person stir the mixture with a straw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 75.0], "base_se_norm": [-0.43828314542770386, -0.35186517238616943], "gt_se_raw": [64.0, 74.0], "gt_se_norm": [-0.44692498445510864, -0.3605070114135742], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006258.pt"}
{"idx": 6260, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8O5vcM_00t0.mp4", "duration": 231.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person start blowing the first bubble?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person start blowing the first bubble?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 96.0], "base_se_norm": [-0.2395218014717102, -0.17038744688034058], "gt_se_raw": [87.0, 97.0], "gt_se_norm": [-0.248163640499115, -0.16174566745758057], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006260.pt"}
{"idx": 6262, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8O5vcM_00t0.mp4", "duration": 231.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the glass is turned upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 187.0 - 195.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the glass is turned upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 187.0 - 195.0 seconds.<|im_end|>\n", "base_se_raw": [187.0, 195.0], "base_se_norm": [0.616016149520874, 0.6851505041122437], "gt_se_raw": [192.0, 208.0], "gt_se_norm": [0.6592251062393188, 0.7974938154220581], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006262.pt"}
{"idx": 6264, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RghziC5GME.mp4", "duration": 204.1622888888889, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy gets angry and crosses his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy gets angry and crosses his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 49.0], "base_se_norm": [-0.5885626077651978, -0.5199897289276123], "gt_se_raw": [42.0, 47.0], "gt_se_norm": [-0.5885626077651978, -0.5395820140838623], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006264.pt"}
{"idx": 6266, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RghziC5GME.mp4", "duration": 204.1622888888889, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the girl shows the boy an instant replay on the TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the girl shows the boy an instant replay on the TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 168.0], "base_se_norm": [0.24410831928253174, 0.645749568939209], "gt_se_raw": [127.0, 133.0], "gt_se_norm": [0.24410831928253174, 0.3028850555419922], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006266.pt"}
{"idx": 6268, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zZNygI-T1_c.mp4", "duration": 303.5949583333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the leather jacket hand a new phone to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the leather jacket hand a new phone to the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.9538859128952026, -0.9275350570678711], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.9472982287406921, -0.9341227412223816], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006268.pt"}
{"idx": 6270, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zZNygI-T1_c.mp4", "duration": 303.5949583333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the woman is sitting on a tree stump, looking at three different phones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the woman is sitting on a tree stump, looking at three different phones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 81.0], "base_se_norm": [-0.565210223197937, -0.466394305229187], "gt_se_raw": [66.0, 74.0], "gt_se_norm": [-0.565210223197937, -0.5125083923339844], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006270.pt"}
{"idx": 6272, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zZNygI-T1_c.mp4", "duration": 303.5949583333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman walks away from the man in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 239.0 - 254.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman walks away from the man in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 239.0 - 254.0 seconds.<|im_end|>\n", "base_se_raw": [239.0, 254.0], "base_se_norm": [0.5744661092758179, 0.6732820272445679], "gt_se_raw": [240.0, 248.0], "gt_se_norm": [0.5810538530349731, 0.6337556838989258], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006272.pt"}
{"idx": 6274, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f8PXvqYpGCM.mp4", "duration": 290.2232666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Donald, the new employee, mocks Brenda's way of speaking for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Donald, the new employee, mocks Brenda's way of speaking for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 42.0], "base_se_norm": [-0.772588849067688, -0.7105676531791687], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [-0.738132655620575, -0.7036764025688171], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006274.pt"}
{"idx": 6276, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f8PXvqYpGCM.mp4", "duration": 290.2232666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where Brenda reveals that her aggressive voice is not her real voice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where Brenda reveals that her aggressive voice is not her real voice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 57.0], "base_se_norm": [-0.6967851519584656, -0.60719895362854], "gt_se_raw": [232.0, 251.0], "gt_se_norm": [0.5987691879272461, 0.7297028303146362], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006276.pt"}
{"idx": 6278, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Lk8pXaDkh8.mp4", "duration": 249.2, "prompt_text": "Please find the visual event described by the sentence 'When does Mario use the F.L.U.D.D. to spray water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Mario use the F.L.U.D.D. to spray water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.9598715901374817, -0.9357945322990417], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.9518458843231201, -0.9277688264846802], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006278.pt"}
{"idx": 6280, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Lk8pXaDkh8.mp4", "duration": 249.2, "prompt_text": "Please find the visual event described by the sentence 'At what point does Hero defeat Little Mac?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 197.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Hero defeat Little Mac?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 197.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [197.0, 201.0], "base_se_norm": [0.581059455871582, 0.6131621599197388], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.6869983673095703, -0.6709470152854919], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006280.pt"}
{"idx": 6282, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Lk8pXaDkh8.mp4", "duration": 249.2, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boxer in the pink outfit fights a pink dragon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boxer in the pink outfit fights a pink dragon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 54.0], "base_se_norm": [-0.6147672533988953, -0.5666131973266602], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.5264847278594971, -0.49438202381134033], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006282.pt"}
{"idx": 6284, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oFV9ayoss_o.mp4", "duration": 204.3043043043043, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a woman with long black hair is writing on a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a woman with long black hair is writing on a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 73.0], "base_se_norm": [-0.3832728862762451, -0.28537970781326294], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [-0.31474769115448, -0.29516905546188354], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006284.pt"}
{"idx": 6286, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oFV9ayoss_o.mp4", "duration": 204.3043043043043, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the black top and leggings stand in front of a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the black top and leggings stand in front of a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 134.0], "base_se_norm": [0.17471826076507568, 0.3117687702178955], "gt_se_raw": [119.0, 133.0], "gt_se_norm": [0.16492891311645508, 0.3019794225692749], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006286.pt"}
{"idx": 6288, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j7rHbHiHyo8.mp4", "duration": 273.73179166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the little pikachu run out of the house to play in the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little pikachu run out of the house to play in the forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [-0.6931302547454834, -0.6492917537689209], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [-0.6931302547454834, -0.6712110042572021], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006288.pt"}
{"idx": 6290, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j7rHbHiHyo8.mp4", "duration": 273.73179166666665, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the scene where the pikachu is forced to battle another pikachu?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the scene where the pikachu is forced to battle another pikachu?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 135.0], "base_se_norm": [-0.12322932481765747, -0.013633012771606445], "gt_se_raw": [237.0, 252.0], "gt_se_norm": [0.7316220998764038, 0.8412184715270996], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006290.pt"}
{"idx": 6292, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j7rHbHiHyo8.mp4", "duration": 273.73179166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the mother pikachu tells her child the rules.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the mother pikachu tells her child the rules.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 39.0], "base_se_norm": [-0.8538715839385986, -0.7150495052337646], "gt_se_raw": [23.0, 39.0], "gt_se_norm": [-0.8319522738456726, -0.7150495052337646], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006292.pt"}
{"idx": 6294, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6M_WjPuZIeA.mp4", "duration": 258.8753222222222, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the hands are folding the yellow paper to make a boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 215.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the hands are folding the yellow paper to make a boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 215.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 215.0], "base_se_norm": [-0.6832451820373535, 0.6610311269760132], "gt_se_raw": [41.0, 218.0], "gt_se_norm": [-0.6832451820373535, 0.6842083930969238], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006294.pt"}
{"idx": 6296, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6M_WjPuZIeA.mp4", "duration": 258.8753222222222, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the clown enlarges the paper boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 220.0 - 236.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the clown enlarges the paper boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 220.0 - 236.0 seconds.<|im_end|>\n", "base_se_raw": [220.0, 236.0], "base_se_norm": [0.6996598243713379, 0.8232713937759399], "gt_se_raw": [229.0, 230.0], "gt_se_norm": [0.7691913843154907, 0.7769170999526978], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006296.pt"}
{"idx": 6298, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/38y_1EWIE9I.mp4", "duration": 281.4, "prompt_text": "Please find the visual event described by the sentence 'When can we see the old lady trying to get her snack out of the vending machine by force?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the old lady trying to get her snack out of the vending machine by force?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 62.0], "base_se_norm": [-0.7512437701225281, -0.559346079826355], "gt_se_raw": [49.0, 57.0], "gt_se_norm": [-0.6517412662506104, -0.5948827266693115], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006298.pt"}
{"idx": 6300, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/38y_1EWIE9I.mp4", "duration": 281.4, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the young man takes the cookie and eats it, surprising the old lady.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the young man takes the cookie and eats it, surprising the old lady.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 160.0], "base_se_norm": [0.08031272888183594, 0.13717126846313477], "gt_se_raw": [106.0, 110.0], "gt_se_norm": [-0.24662399291992188, -0.21819472312927246], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006300.pt"}
{"idx": 6302, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/38y_1EWIE9I.mp4", "duration": 281.4, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the old lady realizes her own cookies were in her bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 196.0 - 205.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the old lady realizes her own cookies were in her bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 196.0 - 205.0 seconds.<|im_end|>\n", "base_se_raw": [196.0, 205.0], "base_se_norm": [0.39303481578826904, 0.457000732421875], "gt_se_raw": [197.0, 205.0], "gt_se_norm": [0.40014219284057617, 0.457000732421875], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006302.pt"}
{"idx": 6304, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pt97zl_gwnk.mp4", "duration": 290.8739166666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where they are pouring the conditioner into the measuring cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where they are pouring the conditioner into the measuring cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 65.0], "base_se_norm": [-0.6080776453018188, -0.5530710220336914], "gt_se_raw": [59.0, 66.0], "gt_se_norm": [-0.5943259596824646, -0.5461951494216919], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006304.pt"}
{"idx": 6306, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pt97zl_gwnk.mp4", "duration": 290.8739166666667, "prompt_text": "Please find the visual event described by the sentence 'When are they mixing the conditioner and cornstarch together in the bowl with their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are they mixing the conditioner and cornstarch together in the bowl with their hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 156.0], "base_se_norm": [-0.030507802963256836, 0.07262969017028809], "gt_se_raw": [129.0, 150.0], "gt_se_norm": [-0.11301779747009277, 0.03137469291687012], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006306.pt"}
{"idx": 6308, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/amsqeYOk--w.mp4", "duration": 314.75, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the book 'No' by David Walsh, PhD is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the book 'No' by David Walsh, PhD is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [-0.669579029083252, -0.6378077864646912], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [-0.669579029083252, -0.6378077864646912], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006308.pt"}
{"idx": 6310, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/amsqeYOk--w.mp4", "duration": 314.75, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the girl in the pink long-sleeved shirt gets a second marshmallow as a reward for waiting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the girl in the pink long-sleeved shirt gets a second marshmallow as a reward for waiting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [-0.35822081565856934, -0.33280378580093384], "gt_se_raw": [133.0, 136.0], "gt_se_norm": [-0.15488481521606445, -0.13582205772399902], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006310.pt"}
{"idx": 6312, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/amsqeYOk--w.mp4", "duration": 314.75, "prompt_text": "Please find the visual event described by the sentence 'At what point does the parenting expert emphasize that consistency from parents is crucial for teaching self-discipline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the parenting expert emphasize that consistency from parents is crucial for teaching self-discipline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 95.0], "base_se_norm": [-0.44082605838775635, -0.3963463306427002], "gt_se_raw": [210.0, 217.0], "gt_se_norm": [0.3343924283981323, 0.3788721561431885], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006312.pt"}
{"idx": 6314, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFu4wn-s-eg.mp4", "duration": 207.47400000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boys place silver dollars into the worn shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boys place silver dollars into the worn shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 68.0], "base_se_norm": [-0.7011673450469971, -0.3444961905479431], "gt_se_raw": [59.0, 68.0], "gt_se_norm": [-0.4312540292739868, -0.3444961905479431], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006314.pt"}
{"idx": 6316, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFu4wn-s-eg.mp4", "duration": 207.47400000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man kneel down and pray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man kneel down and pray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 135.0], "base_se_norm": [0.14713168144226074, 0.3013678789138794], "gt_se_raw": [140.0, 152.0], "gt_se_norm": [0.3495666980743408, 0.4652438163757324], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006316.pt"}
{"idx": 6318, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gAaWm6-4dro.mp4", "duration": 325.1915333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two presenters standing together in the studio at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two presenters standing together in the studio at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 21.0], "base_se_norm": [-1.0, -0.8708453178405762], "gt_se_raw": [0.0, 25.0], "gt_se_norm": [-1.0, -0.8462444543838501], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006318.pt"}
{"idx": 6320, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gAaWm6-4dro.mp4", "duration": 325.1915333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the specialist demonstrates the wrong way to guide by grabbing the woman's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the specialist demonstrates the wrong way to guide by grabbing the woman's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 80.0], "base_se_norm": [-0.5264329314231873, -0.5079822540283203], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.6617377996444702, -0.643287181854248], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006320.pt"}
{"idx": 6322, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gAaWm6-4dro.mp4", "duration": 325.1915333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two women walking together on a snowy sidewalk beside a busy road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 279.0 - 295.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women walking together on a snowy sidewalk beside a busy road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 279.0 - 295.0 seconds.<|im_end|>\n", "base_se_raw": [279.0, 295.0], "base_se_norm": [0.715911865234375, 0.8143154382705688], "gt_se_raw": [280.0, 285.0], "gt_se_norm": [0.7220621109008789, 0.7528132200241089], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006322.pt"}
{"idx": 6324, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/itHff9ErfpQ.mp4", "duration": 214.84796666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where water is added to the glue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where water is added to the glue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 60.0], "base_se_norm": [-0.5066278576850891, -0.44146549701690674], "gt_se_raw": [54.0, 60.0], "gt_se_norm": [-0.49731892347335815, -0.44146549701690674], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006324.pt"}
{"idx": 6326, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/itHff9ErfpQ.mp4", "duration": 214.84796666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the borax solution is added to the main bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the borax solution is added to the main bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 130.0], "base_se_norm": [0.13568687438964844, 0.210158109664917], "gt_se_raw": [122.0, 129.0], "gt_se_norm": [0.13568687438964844, 0.20084917545318604], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006326.pt"}
{"idx": 6328, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ndA4YL-bBAQ.mp4", "duration": 328.88, "prompt_text": "Please find the visual event described by the sentence 'When can we see Betty Boop adding baking powder to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Betty Boop adding baking powder to the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 40.0], "base_se_norm": [-0.8236438632011414, -0.7567501664161682], "gt_se_raw": [30.0, 40.0], "gt_se_norm": [-0.8175626397132874, -0.7567501664161682], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006328.pt"}
{"idx": 6330, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ndA4YL-bBAQ.mp4", "duration": 328.88, "prompt_text": "Please find the visual event described by the sentence 'At what time does the dog get dizzy and spin around after chasing the fly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the dog get dizzy and spin around after chasing the fly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 94.0], "base_se_norm": [-0.47093164920806885, -0.42836296558380127], "gt_se_raw": [87.0, 96.0], "gt_se_norm": [-0.47093164920806885, -0.4162004590034485], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006330.pt"}
{"idx": 6332, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ndA4YL-bBAQ.mp4", "duration": 328.88, "prompt_text": "Please find the visual event described by the sentence 'When is the dog seen with a bucket stuck on its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 277.0 - 284.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog seen with a bucket stuck on its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 277.0 - 284.0 seconds.<|im_end|>\n", "base_se_raw": [277.0, 284.0], "base_se_norm": [0.6845049858093262, 0.7270736694335938], "gt_se_raw": [276.0, 283.0], "gt_se_norm": [0.6784237623214722, 0.7209924459457397], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006332.pt"}
{"idx": 6334, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOxOgGkAjS8.mp4", "duration": 275.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of making the batter by mixing all the ingredients together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of making the batter by mixing all the ingredients together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 84.0], "base_se_norm": [-0.6081277132034302, -0.39042091369628906], "gt_se_raw": [54.0, 83.0], "gt_se_norm": [-0.6081277132034302, -0.39767777919769287], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006334.pt"}
{"idx": 6336, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOxOgGkAjS8.mp4", "duration": 275.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the sauce is simmering in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 217.0 - 227.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the sauce is simmering in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 217.0 - 227.0 seconds.<|im_end|>\n", "base_se_raw": [217.0, 227.0], "base_se_norm": [0.5747460126876831, 0.6473149061203003], "gt_se_raw": [212.0, 228.0], "gt_se_norm": [0.538461446762085, 0.6545717716217041], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006336.pt"}
{"idx": 6338, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RpJd0HeaPHo.mp4", "duration": 247.31373333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the speaker is putting on her makeup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the speaker is putting on her makeup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 58.0], "base_se_norm": [-0.5875684022903442, -0.5309601426124573], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.5633077025413513, -0.5390470027923584], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006338.pt"}
{"idx": 6340, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RpJd0HeaPHo.mp4", "duration": 247.31373333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the speaker addressing a large audience at the 'WE DAY' event.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the speaker addressing a large audience at the 'WE DAY' event.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.9029572606086731, -0.854435920715332], "gt_se_raw": [176.0, 178.0], "gt_se_norm": [0.4232933521270752, 0.439467191696167], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006340.pt"}
{"idx": 6342, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RpJd0HeaPHo.mp4", "duration": 247.31373333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where the speaker is walking towards the stage in a dark theater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 180.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where the speaker is walking towards the stage in a dark theater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 180.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [180.0, 188.0], "base_se_norm": [0.45564091205596924, 0.5203361511230469], "gt_se_raw": [181.0, 183.0], "gt_se_norm": [0.46372783184051514, 0.47990167140960693], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006342.pt"}
{"idx": 6344, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jX9YVuZ8hZc.mp4", "duration": 265.09816666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the man's initial reaction when he hears a strange noise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man's initial reaction when he hears a strange noise.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 60.0], "base_se_norm": [-0.6529587507247925, -0.547337532043457], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [-0.7359468936920166, -0.6906806230545044], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006344.pt"}
{"idx": 6346, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jX9YVuZ8hZc.mp4", "duration": 265.09816666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man start walking down the dark hallway holding a mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start walking down the dark hallway holding a mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 115.0], "base_se_norm": [-0.1852075457572937, -0.1323968768119812], "gt_se_raw": [107.0, 114.0], "gt_se_norm": [-0.19275188446044922, -0.1399412751197815], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006346.pt"}
{"idx": 6348, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8toX8ZjCDVA.mp4", "duration": 298.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of chopping the solid chocolate block.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of chopping the solid chocolate block.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 26.0], "base_se_norm": [-0.9061242938041687, -0.8256593942642212], "gt_se_raw": [14.0, 29.0], "gt_se_norm": [-0.9061242938041687, -0.8055431246757507], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006348.pt"}
{"idx": 6350, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8toX8ZjCDVA.mp4", "duration": 298.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the chocolate mixture is being cooled over an ice bath.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the chocolate mixture is being cooled over an ice bath.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 161.0], "base_se_norm": [-0.06124269962310791, 0.07957088947296143], "gt_se_raw": [151.0, 160.0], "gt_se_norm": [0.01251673698425293, 0.07286548614501953], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006350.pt"}
{"idx": 6352, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8toX8ZjCDVA.mp4", "duration": 298.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a spoon is used to take a bite of the dessert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 276.0 - 289.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a spoon is used to take a bite of the dessert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 276.0 - 289.0 seconds.<|im_end|>\n", "base_se_raw": [276.0, 289.0], "base_se_norm": [0.850692868232727, 0.9378632307052612], "gt_se_raw": [277.0, 282.0], "gt_se_norm": [0.857398271560669, 0.890925407409668], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006352.pt"}
{"idx": 6354, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yK17-IATTE8.mp4", "duration": 204.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman with red hair swings down on a rope and kicks the villain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman with red hair swings down on a rope and kicks the villain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [-0.6478643417358398, -0.5793935060501099], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.6380828022956848, -0.6185197234153748], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006354.pt"}
{"idx": 6356, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yK17-IATTE8.mp4", "duration": 204.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the villain electrocuted by the magic from the stone and the sword?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the villain electrocuted by the magic from the stone and the sword?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 147.0], "base_se_norm": [0.10531473159790039, 0.43788719177246094], "gt_se_raw": [108.0, 118.0], "gt_se_norm": [0.05640697479248047, 0.15422236919403076], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006356.pt"}
{"idx": 6358, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yd_1z6OsrE.mp4", "duration": 276.23429166666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men standing outdoors with a city skyline in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men standing outdoors with a city skyline in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.8334746956825256], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, -0.8986367583274841], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006358.pt"}
{"idx": 6360, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yd_1z6OsrE.mp4", "duration": 276.23429166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a man in a red shirt is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a man in a red shirt is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 51.0], "base_se_norm": [-0.7176309823989868, -0.6307482123374939], "gt_se_raw": [39.0, 51.0], "gt_se_norm": [-0.7176309823989868, -0.6307482123374939], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006360.pt"}
{"idx": 6362, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yd_1z6OsrE.mp4", "duration": 276.23429166666665, "prompt_text": "Please find the visual event described by the sentence 'When is the animation of a train approaching an asterisk shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the animation of a train approaching an asterisk shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 153.0], "base_se_norm": [0.07879436016082764, 0.10775530338287354], "gt_se_raw": [149.0, 152.0], "gt_se_norm": [0.07879436016082764, 0.10051512718200684], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006362.pt"}
{"idx": 6364, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6US4AyvEO_A.mp4", "duration": 262.3037083333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the grey shirt reveal the plot of the movie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the grey shirt reveal the plot of the movie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 122.0], "base_se_norm": [-0.2375251054763794, -0.06978058815002441], "gt_se_raw": [70.0, 81.0], "gt_se_norm": [-0.46626758575439453, -0.38239532709121704], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006364.pt"}
{"idx": 6366, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6US4AyvEO_A.mp4", "duration": 262.3037083333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man in the blue shirt smashes the car with a hammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 202.0 - 206.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man in the blue shirt smashes the car with a hammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 202.0 - 206.0 seconds.<|im_end|>\n", "base_se_raw": [202.0, 206.0], "base_se_norm": [0.5401992797851562, 0.5706983804702759], "gt_se_raw": [202.0, 220.0], "gt_se_norm": [0.5401992797851562, 0.6774448156356812], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006366.pt"}
{"idx": 6368, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pXmHP-7FcP0.mp4", "duration": 210.5, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the blue shirt shake hands with the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the blue shirt shake hands with the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.8479809761047363, -0.8194774389266968], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.8384798169136047, -0.8194774389266968], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006368.pt"}
{"idx": 6370, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pXmHP-7FcP0.mp4", "duration": 210.5, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where Chris explains his coaching experience at the YMCA.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where Chris explains his coaching experience at the YMCA.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 72.0], "base_se_norm": [-0.5629453659057617, -0.315914511680603], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [-0.5344418287277222, -0.4869358539581299], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006370.pt"}
{"idx": 6372, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pXmHP-7FcP0.mp4", "duration": 210.5, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man in the patterned shirt writing in a notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man in the patterned shirt writing in a notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 138.0], "base_se_norm": [-0.1923990249633789, 0.31116390228271484], "gt_se_raw": [113.0, 118.0], "gt_se_norm": [0.07363414764404297, 0.12114012241363525], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006372.pt"}
{"idx": 6374, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TCw-iEJoVfY.mp4", "duration": 236.6530833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a reflection of Carlo appears behind the detective, startling him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a reflection of Carlo appears behind the detective, startling him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [-0.7380131483078003, -0.6957571506500244], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.7295619249343872, -0.6957571506500244], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006374.pt"}
{"idx": 6376, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TCw-iEJoVfY.mp4", "duration": 236.6530833333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the detective shoot the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the detective shoot the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 43.0], "base_se_norm": [-0.6535012722015381, -0.6365988254547119], "gt_se_raw": [172.0, 174.0], "gt_se_norm": [0.4536045789718628, 0.47050702571868896], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006376.pt"}
{"idx": 6378, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HHsmRoHE7K0.mp4", "duration": 230.23, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of making the signature beef filling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of making the signature beef filling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 80.0], "base_se_norm": [-0.6177735328674316, -0.30504274368286133], "gt_se_raw": [44.0, 79.0], "gt_se_norm": [-0.6177735328674316, -0.31372976303100586], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006378.pt"}
{"idx": 6380, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HHsmRoHE7K0.mp4", "duration": 230.23, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range where they are making the classic veggie filling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range where they are making the classic veggie filling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 111.0], "base_se_norm": [-0.30504274368286133, -0.035746872425079346], "gt_se_raw": [80.0, 111.0], "gt_se_norm": [-0.30504274368286133, -0.035746872425079346], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006380.pt"}
{"idx": 6382, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HHsmRoHE7K0.mp4", "duration": 230.23, "prompt_text": "Please find the visual event described by the sentence 'At what point are the baked empanadas with 'VEGGIE' and 'CHEESY' labels displayed in a box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 183.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the baked empanadas with 'VEGGIE' and 'CHEESY' labels displayed in a box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 183.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [183.0, 199.0], "base_se_norm": [0.5897146463394165, 0.7287061214447021], "gt_se_raw": [195.0, 200.0], "gt_se_norm": [0.6939582824707031, 0.7373930215835571], "start_token_span": [81, 86], "end_token_span": [88, 93], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006382.pt"}
{"idx": 6384, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nkpKb2y5BLQ.mp4", "duration": 263.36310000000003, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the importance of reading the syllabus is mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the importance of reading the syllabus is mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 135.0], "base_se_norm": [-0.035552024841308594, 0.025200605392456055], "gt_se_raw": [217.0, 223.0], "gt_se_norm": [0.6479150056838989, 0.6934794187545776], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006384.pt"}
{"idx": 6386, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nkpKb2y5BLQ.mp4", "duration": 263.36310000000003, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a speaker advises students to find their professors' offices during the first week.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a speaker advises students to find their professors' offices during the first week.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.908871054649353, -0.8860888481140137], "gt_se_raw": [147.0, 151.0], "gt_se_norm": [0.11632955074310303, 0.14670586585998535], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006386.pt"}
{"idx": 6388, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BDnXdeoiYRE.mp4", "duration": 312.2285833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man first play with a puppy using a green toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man first play with a puppy using a green toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 17.0], "base_se_norm": [-0.9295387864112854, -0.8911054134368896], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.9231332540512085, -0.8975110054016113], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006388.pt"}
{"idx": 6390, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BDnXdeoiYRE.mp4", "duration": 312.2285833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a clip of Sirius Black's death is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a clip of Sirius Black's death is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 124.0], "base_se_norm": [-0.2633602023124695, -0.20571011304855347], "gt_se_raw": [115.0, 119.0], "gt_se_norm": [-0.2633602023124695, -0.23773795366287231], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006390.pt"}
{"idx": 6392, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BDnXdeoiYRE.mp4", "duration": 312.2285833333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a picture of Cameron Diaz appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 195.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a picture of Cameron Diaz appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 195.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [195.0, 198.0], "base_se_norm": [0.2490849494934082, 0.2683016061782837], "gt_se_raw": [195.0, 197.0], "gt_se_norm": [0.2490849494934082, 0.261896014213562], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006392.pt"}
{"idx": 6394, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OIDEGN4Js40.mp4", "duration": 242.1585777777778, "prompt_text": "Please find the visual event described by the sentence 'When can we see Obama calling someone to find a job?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Obama calling someone to find a job?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 51.0], "base_se_norm": [-0.6448608636856079, -0.5787883996963501], "gt_se_raw": [43.0, 51.0], "gt_se_norm": [-0.6448608636856079, -0.5787883996963501], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006394.pt"}
{"idx": 6396, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OIDEGN4Js40.mp4", "duration": 242.1585777777778, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Obama is making a video on his phone using a filter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Obama is making a video on his phone using a filter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 97.0], "base_se_norm": [-0.2484263777732849, -0.19887208938598633], "gt_se_raw": [91.0, 97.0], "gt_se_norm": [-0.2484263777732849, -0.19887208938598633], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006396.pt"}
{"idx": 6398, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3U-umc-pE20.mp4", "duration": 258.68, "prompt_text": "Please find the visual event described by the sentence 'When does the man pet the dog before they go for a walk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man pet the dog before they go for a walk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 46.0], "base_se_norm": [-0.6598113775253296, -0.6443482637405396], "gt_se_raw": [44.0, 46.0], "gt_se_norm": [-0.6598113775253296, -0.6443482637405396], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006398.pt"}
{"idx": 6400, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3U-umc-pE20.mp4", "duration": 258.68, "prompt_text": "Please find the visual event described by the sentence 'At what time do the man and woman walk hand in hand along the shoreline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do the man and woman walk hand in hand along the shoreline?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 118.0], "base_se_norm": [-0.13406521081924438, -0.08767586946487427], "gt_se_raw": [113.0, 116.0], "gt_se_norm": [-0.12633365392684937, -0.1031389832496643], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006400.pt"}
{"idx": 6402, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3U-umc-pE20.mp4", "duration": 258.68, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the pink-haired woman is about to take a photo with her camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the pink-haired woman is about to take a photo with her camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 93.0], "base_se_norm": [-0.3582804799079895, -0.28096485137939453], "gt_se_raw": [83.0, 87.0], "gt_se_norm": [-0.3582804799079895, -0.3273542523384094], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006402.pt"}
{"idx": 6404, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzRkHkssdME.mp4", "duration": 209.67613333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see muddy, turbulent water flowing rapidly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see muddy, turbulent water flowing rapidly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [-0.5803050994873047, -0.5230740308761597], "gt_se_raw": [43.0, 50.0], "gt_se_norm": [-0.5898436307907104, -0.5230740308761597], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006404.pt"}
{"idx": 6406, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzRkHkssdME.mp4", "duration": 209.67613333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the shot of a single diatom being introduced and examined up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot of a single diatom being introduced and examined up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 96.0], "base_se_norm": [-0.20830285549163818, -0.08430206775665283], "gt_se_raw": [82.0, 96.0], "gt_se_norm": [-0.21784132719039917, -0.08430206775665283], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006406.pt"}
{"idx": 6408, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3XGUnjDRUQ.mp4", "duration": 267.267, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a turtle stuck in a net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a turtle stuck in a net?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9775505661964417], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9925168752670288, -0.9775505661964417], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006408.pt"}
{"idx": 6410, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3XGUnjDRUQ.mp4", "duration": 267.267, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the carcass of a seabird, filled with plastic, is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the carcass of a seabird, filled with plastic, is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 41.0], "base_se_norm": [-0.7605390548706055, -0.6931906938552856], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.7605390548706055, -0.6931906938552856], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006410.pt"}
{"idx": 6412, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3XGUnjDRUQ.mp4", "duration": 267.267, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when a group of dolphins is shown leaping out of the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when a group of dolphins is shown leaping out of the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 140.0], "base_se_norm": [0.0251920223236084, 0.04764151573181152], "gt_se_raw": [165.0, 170.0], "gt_se_norm": [0.2347203493118286, 0.2721360921859741], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006412.pt"}
{"idx": 6414, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pPdevJTGAYY.mp4", "duration": 272.12, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range of the first explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range of the first explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.9338527321815491, -0.9118036031723022], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.5663677453994751, -0.544318675994873], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006414.pt"}
{"idx": 6416, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pPdevJTGAYY.mp4", "duration": 272.12, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a man heating the bottom of the flask with a blue flame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 186.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a man heating the bottom of the flask with a blue flame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 186.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [186.0, 193.0], "base_se_norm": [0.36704397201538086, 0.41849184036254883], "gt_se_raw": [196.0, 202.0], "gt_se_norm": [0.4405409097671509, 0.48463916778564453], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006416.pt"}
{"idx": 6418, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/88tK5c0wgH4.mp4", "duration": 219.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the animation illustrating the transition from a liquid to a gas.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the animation illustrating the transition from a liquid to a gas.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 46.0], "base_se_norm": [-0.7628458738327026, -0.5804195404052734], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.6533900499343872, -0.6169048547744751], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006418.pt"}
{"idx": 6420, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/88tK5c0wgH4.mp4", "duration": 219.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when plasma, the fourth state of matter, is being visually explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when plasma, the fourth state of matter, is being visually explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.5895408987998962, -0.5256916880607605], "gt_se_raw": [46.0, 52.0], "gt_se_norm": [-0.5804195404052734, -0.5256916880607605], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006420.pt"}
{"idx": 6422, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/88tK5c0wgH4.mp4", "duration": 219.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the avalanche begins to cascade down the mountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 195.0 - 202.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the avalanche begins to cascade down the mountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 195.0 - 202.0 seconds.<|im_end|>\n", "base_se_raw": [195.0, 202.0], "base_se_norm": [0.7786561250686646, 0.8425053358078003], "gt_se_raw": [195.0, 200.0], "gt_se_norm": [0.7786561250686646, 0.8242627382278442], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006422.pt"}
{"idx": 6424, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B1yJuGQOUaY.mp4", "duration": 203.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man using the steam wand to froth milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man using the steam wand to froth milk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.8817734122276306, -0.802955687046051], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.871921181678772, -0.8226600885391235], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006424.pt"}
{"idx": 6426, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B1yJuGQOUaY.mp4", "duration": 203.0, "prompt_text": "Please find the visual event described by the sentence 'When does the barista pull the lever to extract the espresso shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the barista pull the lever to extract the espresso shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 70.0], "base_se_norm": [-0.3497536778450012, -0.3103448152542114], "gt_se_raw": [66.0, 68.0], "gt_se_norm": [-0.3497536778450012, -0.3300492763519287], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006426.pt"}
{"idx": 6428, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OPdbdjctx2I.mp4", "duration": 270.4702, "prompt_text": "Please find the visual event described by the sentence 'When does Kevin Hart talk about his palms being clammy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Kevin Hart talk about his palms being clammy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 40.0], "base_se_norm": [-0.7855586409568787, -0.7042188048362732], "gt_se_raw": [31.0, 37.0], "gt_se_norm": [-0.7707695960998535, -0.7264024019241333], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006428.pt"}
{"idx": 6430, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OPdbdjctx2I.mp4", "duration": 270.4702, "prompt_text": "Please find the visual event described by the sentence 'At what point is the roller coaster ascending the very first steep, vertical hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the roller coaster ascending the very first steep, vertical hill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 168.0], "base_se_norm": [0.013050556182861328, 0.2422809600830078], "gt_se_raw": [136.0, 147.0], "gt_se_norm": [0.005656003952026367, 0.08699584007263184], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006430.pt"}
{"idx": 6432, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OPdbdjctx2I.mp4", "duration": 270.4702, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Jimmy Fallon jokes about riding the roller coaster one more time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Jimmy Fallon jokes about riding the roller coaster one more time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 68.0], "base_se_norm": [-0.6154844760894775, -0.4971719980239868], "gt_se_raw": [222.0, 228.0], "gt_se_norm": [0.6415855884552002, 0.6859527826309204], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006432.pt"}
{"idx": 6434, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y8v2hvtM9_c.mp4", "duration": 227.88, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot where the two men are looking at each other intently.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot where the two men are looking at each other intently.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 83.0], "base_se_norm": [-0.7367035150527954, -0.2715464234352112], "gt_se_raw": [30.0, 46.0], "gt_se_norm": [-0.7367035150527954, -0.5962787866592407], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006434.pt"}
{"idx": 6436, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y8v2hvtM9_c.mp4", "duration": 227.88, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when one man shoots the other in the back of the head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when one man shoots the other in the back of the head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 148.0], "base_se_norm": [0.2550464868545532, 0.29892921447753906], "gt_se_raw": [146.0, 149.0], "gt_se_norm": [0.2813761234283447, 0.30770576000213623], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006436.pt"}
{"idx": 6438, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_wsnc8IdCQ.mp4", "duration": 255.24, "prompt_text": "Please find the visual event described by the sentence 'When does the video show clips of Stephen Murray as a professional BMX rider before his accident?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show clips of Stephen Murray as a professional BMX rider before his accident?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.8589562773704529, -0.7962701916694641], "gt_se_raw": [17.0, 26.0], "gt_se_norm": [-0.8667920231819153, -0.7962701916694641], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006438.pt"}
{"idx": 6440, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_wsnc8IdCQ.mp4", "duration": 255.24, "prompt_text": "Please find the visual event described by the sentence 'When is Stephen's father, Jeff Murray, speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Stephen's father, Jeff Murray, speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 118.0], "base_se_norm": [-0.2712740898132324, -0.07538002729415894], "gt_se_raw": [93.0, 119.0], "gt_se_norm": [-0.2712740898132324, -0.06754428148269653], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006440.pt"}
{"idx": 6442, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_wsnc8IdCQ.mp4", "duration": 255.24, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the on-screen keyboard is used to type.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the on-screen keyboard is used to type.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 172.0], "base_se_norm": [0.30857229232788086, 0.3477511405944824], "gt_se_raw": [168.0, 173.0], "gt_se_norm": [0.31640803813934326, 0.3555868864059448], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006442.pt"}
{"idx": 6444, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5itl-NItMdI.mp4", "duration": 270.20353687020355, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when sugar is added to the mashed strawberries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when sugar is added to the mashed strawberries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [-0.49667567014694214, -0.47447019815444946], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.49667567014694214, -0.481872022151947], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006444.pt"}
{"idx": 6446, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5itl-NItMdI.mp4", "duration": 270.20353687020355, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when milk is poured into the bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 221.0 - 231.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when milk is poured into the bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 221.0 - 231.0 seconds.<|im_end|>\n", "base_se_raw": [221.0, 231.0], "base_se_norm": [0.6358039379119873, 0.7098222970962524], "gt_se_raw": [222.0, 230.0], "gt_se_norm": [0.6432058811187744, 0.7024204730987549], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006446.pt"}
{"idx": 6448, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wj8XXvD4kGE.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'When can we see the dry ingredients being whisked together in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the dry ingredients being whisked together in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [-0.6669996976852417, -0.585999608039856], "gt_se_raw": [37.0, 46.0], "gt_se_norm": [-0.6669996976852417, -0.585999608039856], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006448.pt"}
{"idx": 6450, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wj8XXvD4kGE.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person start kneading the dough with their gloved hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person start kneading the dough with their gloved hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 80.0], "base_se_norm": [-0.37899935245513916, -0.2799992561340332], "gt_se_raw": [69.0, 80.0], "gt_se_norm": [-0.37899935245513916, -0.2799992561340332], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006450.pt"}
{"idx": 6452, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wj8XXvD4kGE.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the baked scones are being taken out of the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the baked scones are being taken out of the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 164.0], "base_se_norm": [0.4310014247894287, 0.476001501083374], "gt_se_raw": [159.0, 164.0], "gt_se_norm": [0.4310014247894287, 0.476001501083374], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006452.pt"}
{"idx": 6454, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M-H8mtPWeKw.mp4", "duration": 238.17126666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the thawed peas, carrots, and corn put into the mix?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the thawed peas, carrots, and corn put into the mix?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 68.0], "base_se_norm": [-0.5297501683235168, -0.42898231744766235], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.5129555463790894, -0.48776358366012573], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006454.pt"}
{"idx": 6456, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M-H8mtPWeKw.mp4", "duration": 238.17126666666667, "prompt_text": "Please find the visual event described by the sentence 'Can you show me the moment soy sauce is poured over the rice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you show me the moment soy sauce is poured over the rice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 136.0], "base_se_norm": [0.10844600200653076, 0.1420353651046753], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [0.1420353651046753, 0.158829927444458], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006456.pt"}
{"idx": 6458, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7gquYRxLMFI.mp4", "duration": 205.45524999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me when the grapes are being harvested by hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the grapes are being harvested by hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 62.0], "base_se_norm": [-0.5035415291786194, -0.39646226167678833], "gt_se_raw": [52.0, 61.0], "gt_se_norm": [-0.49380701780319214, -0.4061967134475708], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006458.pt"}
{"idx": 6460, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7gquYRxLMFI.mp4", "duration": 205.45524999999998, "prompt_text": "Please find the visual event described by the sentence 'When are bottles being loaded into a cage for fermentation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are bottles being loaded into a cage for fermentation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 138.0], "base_se_norm": [0.24601340293884277, 0.34335827827453613], "gt_se_raw": [130.0, 138.0], "gt_se_norm": [0.26548242568969727, 0.34335827827453613], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006460.pt"}
{"idx": 6462, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7gquYRxLMFI.mp4", "duration": 205.45524999999998, "prompt_text": "Please find the visual event described by the sentence 'When are workers sorting grapes on a conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are workers sorting grapes on a conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9707965850830078, -0.9123896956443787], "gt_se_raw": [84.0, 92.0], "gt_se_norm": [-0.18230366706848145, -0.10442781448364258], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006462.pt"}
{"idx": 6464, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uTDkM7wUx-A.mp4", "duration": 278.0, "prompt_text": "Please find the visual event described by the sentence 'When does a spider crawl across a woman's face as she lies in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a spider crawl across a woman's face as she lies in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 92.0], "base_se_norm": [-0.3884891867637634, -0.33812952041625977], "gt_se_raw": [91.0, 92.0], "gt_se_norm": [-0.34532374143600464, -0.33812952041625977], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006464.pt"}
{"idx": 6466, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uTDkM7wUx-A.mp4", "duration": 278.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a biplane is taxiing and turning on the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 256.0 - 264.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a biplane is taxiing and turning on the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 256.0 - 264.0 seconds.<|im_end|>\n", "base_se_raw": [256.0, 264.0], "base_se_norm": [0.8417266607284546, 0.8992805480957031], "gt_se_raw": [257.0, 263.0], "gt_se_norm": [0.8489208221435547, 0.892086386680603], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006466.pt"}
{"idx": 6468, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TWfph3iNC-k.mp4", "duration": 209.1089, "prompt_text": "Please find the visual event described by the sentence 'When is the man preparing for his jump by putting on his helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man preparing for his jump by putting on his helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.9234848618507385], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9904356002807617, -0.9234848618507385], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006468.pt"}
{"idx": 6470, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TWfph3iNC-k.mp4", "duration": 209.1089, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the wingsuit flyer is soaring over a large body of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the wingsuit flyer is soaring over a large body of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 57.0], "base_se_norm": [-0.5600378513336182, -0.45482951402664185], "gt_se_raw": [46.0, 57.0], "gt_se_norm": [-0.5600378513336182, -0.45482951402664185], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006470.pt"}
{"idx": 6472, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TWfph3iNC-k.mp4", "duration": 209.1089, "prompt_text": "Please find the visual event described by the sentence 'Find the clip showing the man on the ground, laughing happily after his successful flight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 178.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip showing the man on the ground, laughing happily after his successful flight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 178.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 178.0], "base_se_norm": [0.6737689971923828, 0.7024621963500977], "gt_se_raw": [175.0, 178.0], "gt_se_norm": [0.6737689971923828, 0.7024621963500977], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006472.pt"}
{"idx": 6474, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yjikAEjCPEI.mp4", "duration": 282.8158666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the part where the man uses his cupped hands to create a harmonica-like sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the part where the man uses his cupped hands to create a harmonica-like sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 144.0], "base_se_norm": [-0.13017606735229492, 0.01833045482635498], "gt_se_raw": [123.0, 128.0], "gt_se_norm": [-0.13017606735229492, -0.0948173999786377], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006474.pt"}
{"idx": 6476, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yjikAEjCPEI.mp4", "duration": 282.8158666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is singing the phrase 'Got me feeling so lonesome'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is singing the phrase 'Got me feeling so lonesome'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 144.0], "base_se_norm": [-0.10188913345336914, 0.01833045482635498], "gt_se_raw": [146.0, 148.0], "gt_se_norm": [0.03247392177581787, 0.04661738872528076], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006476.pt"}
{"idx": 6478, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jzF_y039slk.mp4", "duration": 239.69779166666666, "prompt_text": "Please find the visual event described by the sentence 'When is the man playing the cello on the beach during sunset?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man playing the cello on the beach during sunset?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 36.0], "base_se_norm": [-0.9582808017730713, -0.6996217966079712], "gt_se_raw": [6.0, 21.0], "gt_se_norm": [-0.9499369859695435, -0.8247793912887573], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006478.pt"}
{"idx": 6480, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jzF_y039slk.mp4", "duration": 239.69779166666666, "prompt_text": "Please find the visual event described by the sentence 'At what time is a man playing a grand piano on the sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a man playing a grand piano on the sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 54.0], "base_se_norm": [-0.6412148475646973, -0.549432635307312], "gt_se_raw": [65.0, 70.0], "gt_se_norm": [-0.457650363445282, -0.41593116521835327], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006480.pt"}
{"idx": 6482, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jzF_y039slk.mp4", "duration": 239.69779166666666, "prompt_text": "Please find the visual event described by the sentence 'When are both men playing their instruments together in a green, mountainous valley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 221.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are both men playing their instruments together in a green, mountainous valley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 221.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 221.0], "base_se_norm": [0.3683898448944092, 0.8439886569976807], "gt_se_raw": [164.0, 171.0], "gt_se_norm": [0.3683898448944092, 0.42679667472839355], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006482.pt"}
{"idx": 6484, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NLlGopyXT_g.mp4", "duration": 273.4, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the screen shows the HTML source code of the Yahoo! website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the screen shows the HTML source code of the Yahoo! website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 76.0], "base_se_norm": [-0.6561814546585083, -0.4440380334854126], "gt_se_raw": [46.0, 77.0], "gt_se_norm": [-0.663496732711792, -0.4367227554321289], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006484.pt"}
{"idx": 6486, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NLlGopyXT_g.mp4", "duration": 273.4, "prompt_text": "Please find the visual event described by the sentence 'Show me when the user is creating a new blog on the Blogger website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the user is creating a new blog on the Blogger website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 143.0], "base_se_norm": [-0.012435972690582275, 0.04608631134033203], "gt_se_raw": [135.0, 140.0], "gt_se_norm": [-0.012435972690582275, 0.024140477180480957], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006486.pt"}
{"idx": 6488, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fhZo17Pxq1A.mp4", "duration": 207.207, "prompt_text": "Please find the visual event described by the sentence 'When does the rabbit stir the mixture in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the rabbit stir the mixture in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.7876519560813904, -0.7297388315200806], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.7876519560813904, -0.7586954236030579], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006488.pt"}
{"idx": 6490, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fhZo17Pxq1A.mp4", "duration": 207.207, "prompt_text": "Please find the visual event described by the sentence 'At what point is the blue liquid boiling in the green pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the blue liquid boiling in the green pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 82.0], "base_se_norm": [-0.28573840856552124, -0.20852094888687134], "gt_se_raw": [74.0, 82.0], "gt_se_norm": [-0.28573840856552124, -0.20852094888687134], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006490.pt"}
{"idx": 6492, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fhZo17Pxq1A.mp4", "duration": 207.207, "prompt_text": "Please find the visual event described by the sentence 'When can we see the turtle walking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the turtle walking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 167.0], "base_se_norm": [0.16791415214538574, 0.6119146347045898], "gt_se_raw": [141.0, 148.0], "gt_se_norm": [0.3609578609466553, 0.4285231828689575], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006492.pt"}
{"idx": 6494, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t2mU6USTBRE.mp4", "duration": 295.12816666666663, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the man and his group dancing as they jump over the turnstiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the man and his group dancing as they jump over the turnstiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 144.0], "base_se_norm": [-0.05803638696670532, -0.024152815341949463], "gt_se_raw": [140.0, 143.0], "gt_se_norm": [-0.05125969648361206, -0.030929505825042725], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006494.pt"}
{"idx": 6496, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t2mU6USTBRE.mp4", "duration": 295.12816666666663, "prompt_text": "Please find the visual event described by the sentence 'When do we see a close-up of the man's face as he winks at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see a close-up of the man's face as he winks at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [-0.7628149390220642, -0.7357080578804016], "gt_se_raw": [178.0, 180.0], "gt_se_norm": [0.20625555515289307, 0.2198089361190796], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006496.pt"}
{"idx": 6498, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mObyN915RkE.mp4", "duration": 205.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher guide the child to touch the lego blocks representing the short word?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher guide the child to touch the lego blocks representing the short word?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 40.0], "base_se_norm": [-0.7956867218017578, -0.6108318567276001], "gt_se_raw": [23.0, 31.0], "gt_se_norm": [-0.7762283086776733, -0.6983946561813354], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006498.pt"}
{"idx": 6500, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mObyN915RkE.mp4", "duration": 205.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the teacher and child clap their hands together for the long word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the teacher and child clap their hands together for the long word.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 44.0], "base_se_norm": [-0.6108318567276001, -0.5719150304794312], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.6011026501655579, -0.5816442370414734], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006500.pt"}
{"idx": 6502, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mObyN915RkE.mp4", "duration": 205.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher guide the child's hands to touch the long row of Legos?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher guide the child's hands to touch the long row of Legos?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 30.0], "base_se_norm": [-0.7956867218017578, -0.7081239223480225], "gt_se_raw": [32.0, 40.0], "gt_se_norm": [-0.688665509223938, -0.6108318567276001], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006502.pt"}
{"idx": 6504, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ewq5tStHmdk.mp4", "duration": 299.09045555555554, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the couple playfully wears traffic cones on their heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the couple playfully wears traffic cones on their heads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 106.0], "base_se_norm": [-0.3112451434135437, -0.29118430614471436], "gt_se_raw": [104.0, 106.0], "gt_se_norm": [-0.3045582175254822, -0.29118430614471436], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006504.pt"}
{"idx": 6506, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ewq5tStHmdk.mp4", "duration": 299.09045555555554, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man reveal that his new partner is an operating system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man reveal that his new partner is an operating system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 121.0], "base_se_norm": [-0.24437576532363892, -0.19088023900985718], "gt_se_raw": [223.0, 225.0], "gt_se_norm": [0.4911876916885376, 0.5045615434646606], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006506.pt"}
{"idx": 6508, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZREM_AcXHX8.mp4", "duration": 227.5273, "prompt_text": "Please find the visual event described by the sentence 'When does the woman explain the Archer's Paradox?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman explain the Archer's Paradox?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 80.0], "base_se_norm": [-0.43743014335632324, -0.29678767919540405], "gt_se_raw": [64.0, 79.0], "gt_se_norm": [-0.43743014335632324, -0.30557781457901], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006508.pt"}
{"idx": 6510, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZREM_AcXHX8.mp4", "duration": 227.5273, "prompt_text": "Please find the visual event described by the sentence 'Show me when the narrator demonstrates how to bend an arrow with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the narrator demonstrates how to bend an arrow with her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 83.0], "base_se_norm": [-0.29678767919540405, -0.2704172134399414], "gt_se_raw": [80.0, 83.0], "gt_se_norm": [-0.29678767919540405, -0.2704172134399414], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006510.pt"}
{"idx": 6512, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZREM_AcXHX8.mp4", "duration": 227.5273, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is drawing a bow and aiming at the target.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is drawing a bow and aiming at the target.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.7714560031890869, -0.7362953424453735], "gt_se_raw": [38.0, 52.0], "gt_se_norm": [-0.6659741401672363, -0.5429120063781738], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006512.pt"}
{"idx": 6514, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/07d2dXHYb94.mp4", "duration": 245.0, "prompt_text": "Please find the visual event described by the sentence 'When is the trainer teaching a class of puppies in a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the trainer teaching a class of puppies in a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [-0.5265306234359741, -0.4938775300979614], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.5265306234359741, -0.5020408034324646], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006514.pt"}
{"idx": 6516, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/07d2dXHYb94.mp4", "duration": 245.0, "prompt_text": "Please find the visual event described by the sentence 'When does the puppy successfully guide the blind woman through the construction site?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppy successfully guide the blind woman through the construction site?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 199.0], "base_se_norm": [0.4285714626312256, 0.6244897842407227], "gt_se_raw": [176.0, 191.0], "gt_se_norm": [0.436734676361084, 0.5591837167739868], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006516.pt"}
{"idx": 6518, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u6euUDxXlAM.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'When are the two people playfully wrestling for a seat on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two people playfully wrestling for a seat on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 51.0], "base_se_norm": [-0.7512437701225281, -0.6905715465545654], "gt_se_raw": [42.0, 51.0], "gt_se_norm": [-0.7451765537261963, -0.6905715465545654], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006518.pt"}
{"idx": 6520, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u6euUDxXlAM.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'At what point do they sing the 'Mattress Giant' jingle together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 208.0 - 221.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they sing the 'Mattress Giant' jingle together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 208.0 - 221.0 seconds.<|im_end|>\n", "base_se_raw": [208.0, 221.0], "base_se_norm": [0.26198267936706543, 0.340856671333313], "gt_se_raw": [119.0, 126.0], "gt_se_norm": [-0.2780002951622009, -0.23552972078323364], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006520.pt"}
{"idx": 6522, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u6euUDxXlAM.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the two people are dancing during the blooper reel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 304.0 - 321.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the two people are dancing during the blooper reel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 304.0 - 321.0 seconds.<|im_end|>\n", "base_se_raw": [304.0, 321.0], "base_se_norm": [0.8444362878799438, 0.9475791454315186], "gt_se_raw": [301.0, 317.0], "gt_se_norm": [0.8262345790863037, 0.9233101606369019], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006522.pt"}
{"idx": 6524, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggt_6Xb0c48.mp4", "duration": 301.80150000000003, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman use a brush to apply baked powder foundation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman use a brush to apply baked powder foundation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 133.0], "base_se_norm": [-0.21140223741531372, -0.11862599849700928], "gt_se_raw": [119.0, 133.0], "gt_se_norm": [-0.21140223741531372, -0.11862599849700928], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006524.pt"}
{"idx": 6526, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggt_6Xb0c48.mp4", "duration": 301.80150000000003, "prompt_text": "Please find the visual event described by the sentence 'Show the period where the woman is filling in her eyebrows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 215.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period where the woman is filling in her eyebrows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 215.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 215.0], "base_se_norm": [0.27235937118530273, 0.4247775077819824], "gt_se_raw": [193.0, 206.0], "gt_se_norm": [0.2789863348007202, 0.36513566970825195], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006526.pt"}
{"idx": 6528, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BxyXk1sr2io.mp4", "duration": 251.08, "prompt_text": "Please find the visual event described by the sentence 'When does the person fold the paper in half horizontally?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person fold the paper in half horizontally?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 19.0], "base_se_norm": [-0.9362753033638, -0.8486537933349609], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.9442408680915833, -0.9203441143035889], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006528.pt"}
{"idx": 6530, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BxyXk1sr2io.mp4", "duration": 251.08, "prompt_text": "Please find the visual event described by the sentence 'At what point are the colors 'Red', 'Blue', 'Orange', and 'Purple' written on the flaps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the colors 'Red', 'Blue', 'Orange', and 'Purple' written on the flaps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 138.0], "base_se_norm": [-0.02023261785507202, 0.09925127029418945], "gt_se_raw": [124.0, 137.0], "gt_se_norm": [-0.012266993522644043, 0.0912855863571167], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006530.pt"}
{"idx": 6532, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BxyXk1sr2io.mp4", "duration": 251.08, "prompt_text": "Please find the visual event described by the sentence 'When are the profession names written inside the flaps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the profession names written inside the flaps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 199.0], "base_se_norm": [-0.02023261785507202, 0.5851521492004395], "gt_se_raw": [164.0, 198.0], "gt_se_norm": [0.3063565492630005, 0.5771865844726562], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006532.pt"}
{"idx": 6534, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RLmKfXwWQtE.mp4", "duration": 301.42612499999996, "prompt_text": "Please find the visual event described by the sentence 'Show the comparison of the population pyramids of Rwanda, Canada, and Japan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 146.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the comparison of the population pyramids of Rwanda, Canada, and Japan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 146.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [146.0, 155.0], "base_se_norm": [-0.03127175569534302, 0.028444409370422363], "gt_se_raw": [146.0, 155.0], "gt_se_norm": [-0.03127175569534302, 0.028444409370422363], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006534.pt"}
{"idx": 6536, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RLmKfXwWQtE.mp4", "duration": 301.42612499999996, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that explains how Russia's population pyramid was affected by World War II.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 221.0 - 237.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that explains how Russia's population pyramid was affected by World War II.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 221.0 - 237.0 seconds.<|im_end|>\n", "base_se_raw": [221.0, 237.0], "base_se_norm": [0.4663625955581665, 0.5725246667861938], "gt_se_raw": [222.0, 237.0], "gt_se_norm": [0.472997784614563, 0.5725246667861938], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006536.pt"}
{"idx": 6538, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E1dVEAdYuaM.mp4", "duration": 261.6, "prompt_text": "Please find the visual event described by the sentence 'When does the woman first realize the package contains her book manuscript?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman first realize the package contains her book manuscript?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.8776758313179016, -0.8318042755126953], "gt_se_raw": [20.0, 25.0], "gt_se_norm": [-0.8470947742462158, -0.8088685274124146], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006538.pt"}
{"idx": 6540, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E1dVEAdYuaM.mp4", "duration": 261.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman run through the garden to find the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman run through the garden to find the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 87.0], "base_se_norm": [-0.5565749406814575, -0.3348624110221863], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [-0.5565749406814575, -0.5030580759048462], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006540.pt"}
{"idx": 6542, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E1dVEAdYuaM.mp4", "duration": 261.6, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the couple kiss under the umbrella in the rain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 244.0 - 248.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the couple kiss under the umbrella in the rain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 244.0 - 248.0 seconds.<|im_end|>\n", "base_se_raw": [244.0, 248.0], "base_se_norm": [0.8654433488845825, 0.8960244655609131], "gt_se_raw": [243.0, 249.0], "gt_se_norm": [0.8577980995178223, 0.9036697149276733], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006542.pt"}
{"idx": 6544, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ta9K22D0o5Q.mp4", "duration": 251.4178333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the shell is hiding inside a slipper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 197.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the shell is hiding inside a slipper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 197.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 197.0], "base_se_norm": [0.4716538190841675, 0.5671124458312988], "gt_se_raw": [185.0, 190.0], "gt_se_norm": [0.4716538190841675, 0.5114282369613647], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006544.pt"}
{"idx": 6546, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ta9K22D0o5Q.mp4", "duration": 251.4178333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the shell attempt to turn a book page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the shell attempt to turn a book page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 124.0], "base_se_norm": [-0.06132352352142334, -0.013594210147857666], "gt_se_raw": [118.0, 122.0], "gt_se_norm": [-0.06132352352142334, -0.029504001140594482], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006546.pt"}
{"idx": 6548, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ka8-Nefp_gk.mp4", "duration": 297.5833333333333, "prompt_text": "Please find the visual event described by the sentence 'When does a hand write the word 'College' on a clipboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a hand write the word 'College' on a clipboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.7177261114120483, -0.6908429265022278], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.7110053300857544, -0.6975637078285217], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006548.pt"}
{"idx": 6550, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ka8-Nefp_gk.mp4", "duration": 297.5833333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two people are sitting on a bench in a park during autumn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two people are sitting on a bench in a park during autumn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 188.0], "base_se_norm": [-0.16662001609802246, 0.2635115385055542], "gt_se_raw": [123.0, 168.0], "gt_se_norm": [-0.17334085702896118, 0.1290954351425171], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006550.pt"}
{"idx": 6552, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ka8-Nefp_gk.mp4", "duration": 297.5833333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are people building with large wooden blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 224.0 - 256.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people building with large wooden blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 224.0 - 256.0 seconds.<|im_end|>\n", "base_se_raw": [224.0, 256.0], "base_se_norm": [0.5054606199264526, 0.7205264568328857], "gt_se_raw": [223.0, 255.0], "gt_se_norm": [0.4987398386001587, 0.7138055562973022], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006552.pt"}
{"idx": 6554, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/phQDinMbmic.mp4", "duration": 219.8863333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the crumpled paper ball is thrown and lands on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the crumpled paper ball is thrown and lands on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 108.0], "base_se_norm": [-0.054056763648986816, -0.01767432689666748], "gt_se_raw": [104.0, 107.0], "gt_se_norm": [-0.054056763648986816, -0.026769936084747314], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006554.pt"}
{"idx": 6556, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/phQDinMbmic.mp4", "duration": 219.8863333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl hold up her notebook to show the boy her drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl hold up her notebook to show the boy her drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 52.0], "base_se_norm": [-0.6088888645172119, -0.5270283818244934], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [-0.39968985319137573, -0.3633074164390564], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006556.pt"}
{"idx": 6558, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5GN90UiImo.mp4", "duration": 314.88123333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man transform, with his shirt ripping apart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man transform, with his shirt ripping apart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 105.0], "base_se_norm": [-0.428355872631073, -0.33308184146881104], "gt_se_raw": [91.0, 105.0], "gt_se_norm": [-0.4220042824745178, -0.33308184146881104], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006558.pt"}
{"idx": 6560, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5GN90UiImo.mp4", "duration": 314.88123333333334, "prompt_text": "Please find the visual event described by the sentence 'When does The Rock Obama rip the man's arm off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 194.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does The Rock Obama rip the man's arm off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 194.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 194.0], "base_se_norm": [0.14963984489440918, 0.23221063613891602], "gt_se_raw": [181.0, 183.0], "gt_se_norm": [0.14963984489440918, 0.16234302520751953], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006560.pt"}
{"idx": 6562, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5GN90UiImo.mp4", "duration": 314.88123333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the woman in the red dress rips off the agent's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 283.0 - 288.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the woman in the red dress rips off the agent's arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 283.0 - 288.0 seconds.<|im_end|>\n", "base_se_raw": [283.0, 288.0], "base_se_norm": [0.7975031137466431, 0.8292611837387085], "gt_se_raw": [283.0, 286.0], "gt_se_norm": [0.7975031137466431, 0.8165580034255981], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006562.pt"}
{"idx": 6564, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MoOXcFZgzhI.mp4", "duration": 218.05116666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a group of young people are sitting in a circle on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a group of young people are sitting in a circle on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [-0.29374372959136963, -0.24788296222686768], "gt_se_raw": [78.0, 82.0], "gt_se_norm": [-0.2845715880393982, -0.24788296222686768], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006564.pt"}
{"idx": 6566, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MoOXcFZgzhI.mp4", "duration": 218.05116666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the scene where a woman in a red sweater is talking to a student in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the scene where a woman in a red sweater is talking to a student in a classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 146.0], "base_se_norm": [0.25658583641052246, 0.3391352891921997], "gt_se_raw": [143.0, 146.0], "gt_se_norm": [0.3116188049316406, 0.3391352891921997], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006566.pt"}
{"idx": 6568, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CN8899eqays.mp4", "duration": 311.97833333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the fake jaw is being painted red.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the fake jaw is being painted red.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 39.0], "base_se_norm": [-0.788446843624115, -0.7499825954437256], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.7820361256599426, -0.756393313407898], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006568.pt"}
{"idx": 6570, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CN8899eqays.mp4", "duration": 311.97833333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the woman apply dark makeup to create sunken eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman apply dark makeup to create sunken eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 74.0], "base_se_norm": [-0.6089471578598022, -0.5256080627441406], "gt_se_raw": [72.0, 86.0], "gt_se_norm": [-0.5384294986724854, -0.44867968559265137], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006570.pt"}
{"idx": 6572, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CN8899eqays.mp4", "duration": 311.97833333333335, "prompt_text": "Please find the visual event described by the sentence 'When is fake blood being applied to the wound on her shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 237.0 - 286.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is fake blood being applied to the wound on her shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 237.0 - 286.0 seconds.<|im_end|>\n", "base_se_raw": [237.0, 286.0], "base_se_norm": [0.5193362236022949, 0.8334606885910034], "gt_se_raw": [274.0, 286.0], "gt_se_norm": [0.7565321922302246, 0.8334606885910034], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006572.pt"}
{"idx": 6574, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKGeJXk2qWQ.mp4", "duration": 234.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the cashier reading a magazine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cashier reading a magazine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [-0.34409940242767334, -0.3100266456604004], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.34409940242767334, -0.3100266456604004], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006574.pt"}
{"idx": 6576, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKGeJXk2qWQ.mp4", "duration": 234.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a stick of butter dropped into the hole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a stick of butter dropped into the hole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 167.0], "base_se_norm": [0.40550124645233154, 0.4225376844406128], "gt_se_raw": [164.0, 167.0], "gt_se_norm": [0.39698314666748047, 0.4225376844406128], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006576.pt"}
{"idx": 6578, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VF9-sEbqDvU.mp4", "duration": 201.61808333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the shell talk on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the shell talk on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 89.0], "base_se_norm": [-0.3056178689002991, -0.1171426773071289], "gt_se_raw": [70.0, 89.0], "gt_se_norm": [-0.3056178689002991, -0.1171426773071289], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006578.pt"}
{"idx": 6580, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VF9-sEbqDvU.mp4", "duration": 201.61808333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the shell is dragging a piece of lint with a hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the shell is dragging a piece of lint with a hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 131.0], "base_se_norm": [0.24988782405853271, 0.2994866371154785], "gt_se_raw": [126.0, 130.0], "gt_se_norm": [0.24988782405853271, 0.28956687450408936], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006580.pt"}
{"idx": 6582, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VF9-sEbqDvU.mp4", "duration": 201.61808333333332, "prompt_text": "Please find the visual event described by the sentence 'What time does the shell introduce itself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What time does the shell introduce itself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 22.0], "base_se_norm": [-0.9900802373886108, -0.7817655801773071], "gt_se_raw": [3.0, 25.0], "gt_se_norm": [-0.9702407717704773, -0.7520064115524292], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006582.pt"}
{"idx": 6584, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BA2yyhbHcCc.mp4", "duration": 309.20920920920923, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see the brine shrimp and the rock shrimp through the microscope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see the brine shrimp and the rock shrimp through the microscope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 81.0], "base_se_norm": [-0.5278277397155762, -0.47608286142349243], "gt_se_raw": [74.0, 81.0], "gt_se_norm": [-0.5213596820831299, -0.47608286142349243], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006584.pt"}
{"idx": 6586, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BA2yyhbHcCc.mp4", "duration": 309.20920920920923, "prompt_text": "Please find the visual event described by the sentence 'When does the girl pour water into the cup to create the microscope lens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 257.0 - 267.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl pour water into the cup to create the microscope lens?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 257.0 - 267.0 seconds.<|im_end|>\n", "base_se_raw": [257.0, 267.0], "base_se_norm": [0.6623049974441528, 0.7269861698150635], "gt_se_raw": [265.0, 267.0], "gt_se_norm": [0.7140499353408813, 0.7269861698150635], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006586.pt"}
{"idx": 6588, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oRmpxPwJFw8.mp4", "duration": 222.55566666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the woman explain the rule for remembering where the forks go?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman explain the rule for remembering where the forks go?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 64.0], "base_se_norm": [-0.4697955846786499, -0.42486298084259033], "gt_se_raw": [39.0, 48.0], "gt_se_norm": [-0.6495258808135986, -0.5686472654342651], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006588.pt"}
{"idx": 6590, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oRmpxPwJFw8.mp4", "duration": 222.55566666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman place the main dinner plate in the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman place the main dinner plate in the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 115.0], "base_se_norm": [-0.056415855884552, 0.03344929218292236], "gt_se_raw": [105.0, 109.0], "gt_se_norm": [-0.056415855884552, -0.0204697847366333], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006590.pt"}
{"idx": 6592, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oRmpxPwJFw8.mp4", "duration": 222.55566666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where she explains her 'rule of thumb' for spacing the place setting from the edge of the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where she explains her 'rule of thumb' for spacing the place setting from the edge of the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 115.0], "base_se_norm": [-0.056415855884552, 0.03344929218292236], "gt_se_raw": [198.0, 207.0], "gt_se_norm": [0.7793301343917847, 0.8602087497711182], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006592.pt"}
{"idx": 6594, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wx9vPv-T51I.mp4", "duration": 216.5079583333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a huge wave is rushing towards the soldiers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a huge wave is rushing towards the soldiers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.7967742085456848, -0.7690616250038147], "gt_se_raw": [128.0, 130.0], "gt_se_norm": [0.1824045181274414, 0.20087957382202148], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006594.pt"}
{"idx": 6596, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wx9vPv-T51I.mp4", "duration": 216.5079583333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a surfer rides a big wave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a surfer rides a big wave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.6674487590789795, -0.6304985880851746], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [-0.6582112312316895, -0.6304985880851746], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006596.pt"}
{"idx": 6598, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xiSIQzwIPzQ.mp4", "duration": 210.2, "prompt_text": "Please find the visual event described by the sentence 'When can we see two people sitting in the middle of a pumpkin patch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two people sitting in the middle of a pumpkin patch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 46.0], "base_se_norm": [-0.923882007598877, -0.5623216032981873], "gt_se_raw": [8.0, 15.0], "gt_se_norm": [-0.923882007598877, -0.8572787642478943], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006598.pt"}
{"idx": 6600, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xiSIQzwIPzQ.mp4", "duration": 210.2, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy faints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy faints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 57.0], "base_se_norm": [-0.4957183599472046, -0.45765936374664307], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.4862036108970642, -0.45765936374664307], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006600.pt"}
{"idx": 6602, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xiSIQzwIPzQ.mp4", "duration": 210.2, "prompt_text": "Please find the visual event described by the sentence 'When does a girl in a yellow coat help a boy go to bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 204.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a girl in a yellow coat help a boy go to bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 204.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 204.0], "base_se_norm": [0.6841104030609131, 0.9410085678100586], "gt_se_raw": [197.0, 205.0], "gt_se_norm": [0.8744053840637207, 0.9505233764648438], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006602.pt"}
{"idx": 6604, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3X1nPsafKdk.mp4", "duration": 208.0, "prompt_text": "Please find the visual event described by the sentence 'When does the puppet say it has been a team with the girl for two years?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppet say it has been a team with the girl for two years?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 66.0], "base_se_norm": [-0.42307692766189575, -0.36538463830947876], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.75, -0.7211538553237915], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006604.pt"}
{"idx": 6606, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3X1nPsafKdk.mp4", "duration": 208.0, "prompt_text": "Please find the visual event described by the sentence 'When does the host's puppet start to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the host's puppet start to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 111.0], "base_se_norm": [-0.019230782985687256, 0.06730771064758301], "gt_se_raw": [158.0, 163.0], "gt_se_norm": [0.5192307233810425, 0.567307710647583], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006606.pt"}
{"idx": 6608, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJJBHDi9axY.mp4", "duration": 324.69103333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the TV test pattern being broadcasted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the TV test pattern being broadcasted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 43.0], "base_se_norm": [-1.0, -0.7351328134536743], "gt_se_raw": [0.0, 44.0], "gt_se_norm": [-1.0, -0.7289731502532959], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006608.pt"}
{"idx": 6610, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJJBHDi9axY.mp4", "duration": 324.69103333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the plaid shirt adjust his tie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the plaid shirt adjust his tie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 118.0], "base_se_norm": [-0.4271477460861206, -0.273155152797699], "gt_se_raw": [114.0, 119.0], "gt_se_norm": [-0.297793984413147, -0.26699548959732056], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006610.pt"}
{"idx": 6612, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJJBHDi9axY.mp4", "duration": 324.69103333333334, "prompt_text": "Please find the visual event described by the sentence 'When does a person with a box on their head walk towards a body on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 251.0 - 282.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a person with a box on their head walk towards a body on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 251.0 - 282.0 seconds.<|im_end|>\n", "base_se_raw": [251.0, 282.0], "base_se_norm": [0.5460851192474365, 0.737035870552063], "gt_se_raw": [250.0, 282.0], "gt_se_norm": [0.5399254560470581, 0.737035870552063], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006612.pt"}
{"idx": 6614, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2is08NsKzkw.mp4", "duration": 256.7565, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of an animation that illustrates the Yellowstone Caldera and the Hot Spot beneath it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of an animation that illustrates the Yellowstone Caldera and the Hot Spot beneath it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 78.0], "base_se_norm": [-0.47031527757644653, -0.3924204707145691], "gt_se_raw": [67.0, 78.0], "gt_se_norm": [-0.47810477018356323, -0.3924204707145691], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006614.pt"}
{"idx": 6616, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2is08NsKzkw.mp4", "duration": 256.7565, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when two elk are seen sparring on a meadow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 198.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when two elk are seen sparring on a meadow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 198.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [198.0, 201.0], "base_se_norm": [0.542317271232605, 0.5656857490539551], "gt_se_raw": [198.0, 201.0], "gt_se_norm": [0.542317271232605, 0.5656857490539551], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006616.pt"}
{"idx": 6618, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFtBjc1dz7w.mp4", "duration": 290.2482888888889, "prompt_text": "Please find the visual event described by the sentence 'When does the video compare the plot of The Lion King to Frozen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video compare the plot of The Lion King to Frozen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 30.0], "base_se_norm": [-0.8759682774543762, -0.7932804226875305], "gt_se_raw": [18.0, 30.0], "gt_se_norm": [-0.8759682774543762, -0.7932804226875305], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006618.pt"}
{"idx": 6620, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFtBjc1dz7w.mp4", "duration": 290.2482888888889, "prompt_text": "Please find the visual event described by the sentence 'Show the part where Scar is singing his villain song, surrounded by green smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where Scar is singing his villain song, surrounded by green smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 135.0], "base_se_norm": [-0.1386684775352478, -0.06976199150085449], "gt_se_raw": [127.0, 135.0], "gt_se_norm": [-0.12488716840744019, -0.06976199150085449], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006620.pt"}
{"idx": 6622, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFtBjc1dz7w.mp4", "duration": 290.2482888888889, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a young Simba is held up on Pride Rock by Rafiki.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a young Simba is held up on Pride Rock by Rafiki.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.7381552457809448, -0.7174832820892334], "gt_se_raw": [204.0, 206.0], "gt_se_norm": [0.40569305419921875, 0.41947436332702637], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006622.pt"}
{"idx": 6624, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCsVT9TKahk.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where she explains why she started wearing baseball hats after her injury.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where she explains why she started wearing baseball hats after her injury.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 60.0], "base_se_norm": [-0.7012944221496582, -0.40258878469467163], "gt_se_raw": [41.0, 62.0], "gt_se_norm": [-0.5917689800262451, -0.38267505168914795], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006624.pt"}
{"idx": 6626, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCsVT9TKahk.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment she mentions being hit by a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment she mentions being hit by a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.9004314541816711, -0.870560884475708], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9303020238876343, -0.9103883504867554], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006626.pt"}
{"idx": 6628, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_TbWcdN-W8o.mp4", "duration": 201.93506666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man walking on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man walking on the sidewalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 43.0], "base_se_norm": [-0.6830664873123169, -0.5741205215454102], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.6830664873123169, -0.6335455775260925], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006628.pt"}
{"idx": 6630, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_TbWcdN-W8o.mp4", "duration": 201.93506666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is he shown putting on his red sneakers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is he shown putting on his red sneakers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 94.0], "base_se_norm": [-0.11852854490280151, -0.06900769472122192], "gt_se_raw": [89.0, 94.0], "gt_se_norm": [-0.11852854490280151, -0.06900769472122192], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006630.pt"}
{"idx": 6632, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_TbWcdN-W8o.mp4", "duration": 201.93506666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man first introduce himself and state his age?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man first introduce himself and state his age?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7523956298828125, -0.7226831316947937], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.7424914836883545, -0.7028747797012329], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006632.pt"}
{"idx": 6634, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AW0jm6i9U3M.mp4", "duration": 298.46483333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the driver hold a pipe up to the passenger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the driver hold a pipe up to the passenger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 44.0], "base_se_norm": [-0.772167444229126, -0.7051578760147095], "gt_se_raw": [31.0, 36.0], "gt_se_norm": [-0.7922703623771667, -0.7587655782699585], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006634.pt"}
{"idx": 6636, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AW0jm6i9U3M.mp4", "duration": 298.46483333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the passenger finally decide to smoke the pipe and asks for it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the passenger finally decide to smoke the pipe and asks for it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 114.0], "base_se_norm": [-0.269595742225647, -0.23609095811843872], "gt_se_raw": [94.0, 98.0], "gt_se_norm": [-0.37011009454727173, -0.3433062434196472], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006636.pt"}
{"idx": 6638, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zucl7HSA1M.mp4", "duration": 252.48556666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a cat in a pirate costume falling over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a cat in a pirate costume falling over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.8970237970352173, -0.8574175834655762], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.889102578163147, -0.8653388023376465], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006638.pt"}
{"idx": 6640, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zucl7HSA1M.mp4", "duration": 252.48556666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat fall into the pond?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat fall into the pond?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 112.0], "base_se_norm": [-0.1603480577468872, -0.112820565700531], "gt_se_raw": [106.0, 108.0], "gt_se_norm": [-0.1603480577468872, -0.1445055603981018], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006640.pt"}
{"idx": 6642, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zucl7HSA1M.mp4", "duration": 252.48556666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the cat attempting to climb the wall to get to the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cat attempting to climb the wall to get to the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 199.0], "base_se_norm": [0.46543025970458984, 0.5763276815414429], "gt_se_raw": [189.0, 199.0], "gt_se_norm": [0.49711525440216064, 0.5763276815414429], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006642.pt"}
{"idx": 6644, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zWHu95io9B4.mp4", "duration": 219.63608333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the egg is being beaten with a whisk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the egg is being beaten with a whisk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.48095959424972534, -0.43542975187301636], "gt_se_raw": [57.0, 62.0], "gt_se_norm": [-0.48095959424972534, -0.43542975187301636], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006644.pt"}
{"idx": 6646, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zWHu95io9B4.mp4", "duration": 219.63608333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the chopped onions and garlic are stirred and fried in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the chopped onions and garlic are stirred and fried in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 132.0], "base_se_norm": [0.12914049625396729, 0.2019883394241333], "gt_se_raw": [124.0, 132.0], "gt_se_norm": [0.12914049625396729, 0.2019883394241333], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006646.pt"}
{"idx": 6648, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UuAUKwAhIjY.mp4", "duration": 233.06616666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a boat being rowed on the foggy river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a boat being rowed on the foggy river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 55.0], "base_se_norm": [-0.9742562174797058, -0.5280309915542603], "gt_se_raw": [37.0, 55.0], "gt_se_norm": [-0.6824935674667358, -0.5280309915542603], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006648.pt"}
{"idx": 6650, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UuAUKwAhIjY.mp4", "duration": 233.06616666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman is giving birth, supported by other women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman is giving birth, supported by other women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 122.0], "base_se_norm": [-0.21052461862564087, 0.0469130277633667], "gt_se_raw": [92.0, 122.0], "gt_se_norm": [-0.21052461862564087, 0.0469130277633667], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006650.pt"}
{"idx": 6652, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UuAUKwAhIjY.mp4", "duration": 233.06616666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where someone is writing with a quill pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where someone is writing with a quill pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 198.0], "base_se_norm": [0.5617883205413818, 0.6990883350372314], "gt_se_raw": [182.0, 198.0], "gt_se_norm": [0.5617883205413818, 0.6990883350372314], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006652.pt"}
{"idx": 6654, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QjqS7jzjX34.mp4", "duration": 285.4184666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Gambit leaps and throws his glowing cards, causing an explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Gambit leaps and throws his glowing cards, causing an explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.7827751040458679, -0.7617533206939697], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [-0.7827751040458679, -0.7687605619430542], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006654.pt"}
{"idx": 6656, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QjqS7jzjX34.mp4", "duration": 285.4184666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does Wolverine jump on Omega Red's back and attack him with his claws?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Wolverine jump on Omega Red's back and attack him with his claws?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 143.0], "base_se_norm": [-0.06102776527404785, 0.002037525177001953], "gt_se_raw": [133.0, 137.0], "gt_se_norm": [-0.0680350661277771, -0.040006041526794434], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006656.pt"}
{"idx": 6658, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/alLd1Bobkf8.mp4", "duration": 260.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'Where in the video is the concept of a student being available for learning visually represented by a doll in an open box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video is the concept of a student being available for learning visually represented by a doll in an open box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 38.0], "base_se_norm": [-0.7618438005447388, -0.7080665826797485], "gt_se_raw": [31.0, 39.0], "gt_se_norm": [-0.7618438005447388, -0.7003841400146484], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006658.pt"}
{"idx": 6660, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/alLd1Bobkf8.mp4", "duration": 260.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where a model of the human brain is used to illustrate that stress inhibits learning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 226.0 - 241.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where a model of the human brain is used to illustrate that stress inhibits learning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 226.0 - 241.0 seconds.<|im_end|>\n", "base_se_raw": [226.0, 241.0], "base_se_norm": [0.736235499382019, 0.8514723777770996], "gt_se_raw": [226.0, 241.0], "gt_se_norm": [0.736235499382019, 0.8514723777770996], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006660.pt"}
{"idx": 6662, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/alLd1Bobkf8.mp4", "duration": 260.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range where the presenter explains the signs of the 'lid being flung open', such as sensory-seeking behaviors and agitation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 205.0 - 225.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range where the presenter explains the signs of the 'lid being flung open', such as sensory-seeking behaviors and agitation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 205.0 - 225.0 seconds.<|im_end|>\n", "base_se_raw": [205.0, 225.0], "base_se_norm": [0.5749039649963379, 0.728553056716919], "gt_se_raw": [205.0, 226.0], "gt_se_norm": [0.5749039649963379, 0.736235499382019], "start_token_span": [82, 87], "end_token_span": [89, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006662.pt"}
{"idx": 6664, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iD2rhdFRehU.mp4", "duration": 291.16587499999997, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a close-up of two glasses being filled with drinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a close-up of two glasses being filled with drinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 83.0], "base_se_norm": [-0.4779607653617859, -0.42987823486328125], "gt_se_raw": [82.0, 83.0], "gt_se_norm": [-0.43674713373184204, -0.42987823486328125], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006664.pt"}
{"idx": 6666, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iD2rhdFRehU.mp4", "duration": 291.16587499999997, "prompt_text": "Please find the visual event described by the sentence 'When does the man, wearing a suit, sit at the head of a long, lavishly set dining table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 220.0 - 223.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man, wearing a suit, sit at the head of a long, lavishly set dining table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 220.0 - 223.0 seconds.<|im_end|>\n", "base_se_raw": [220.0, 223.0], "base_se_norm": [0.5111662149429321, 0.5317729711532593], "gt_se_raw": [220.0, 223.0], "gt_se_norm": [0.5111662149429321, 0.5317729711532593], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006666.pt"}
{"idx": 6668, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsX3ATc3FbA.mp4", "duration": 252.50224999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman with pink and orange hair eating a lollipop at the ticket booth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman with pink and orange hair eating a lollipop at the ticket booth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.9366342425346375, -0.8732684850692749], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.9287135004997253, -0.9128720760345459], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006668.pt"}
{"idx": 6670, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsX3ATc3FbA.mp4", "duration": 252.50224999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man with blue hair dance on a giant piano keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man with blue hair dance on a giant piano keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 129.0], "base_se_norm": [-0.09703773260116577, 0.021773099899291992], "gt_se_raw": [118.0, 129.0], "gt_se_norm": [-0.06535482406616211, 0.021773099899291992], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006670.pt"}
{"idx": 6672, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsX3ATc3FbA.mp4", "duration": 252.50224999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the group dances in white suits in front of a colorful, painted city skyline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the group dances in white suits in front of a colorful, painted city skyline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 168.0], "base_se_norm": [0.1168217658996582, 0.3306812047958374], "gt_se_raw": [146.0, 155.0], "gt_se_norm": [0.1564253568649292, 0.22771191596984863], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006672.pt"}
{"idx": 6674, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgcISkB2duI.mp4", "duration": 282.6666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy put a toy ball into the orange pumpkin bucket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy put a toy ball into the orange pumpkin bucket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [-0.5896226167678833, -0.5542452335357666], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [-0.483490526676178, -0.44811320304870605], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006674.pt"}
{"idx": 6676, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgcISkB2duI.mp4", "duration": 282.6666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman demonstrate the concepts of 'on' and 'in' using the boxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman demonstrate the concepts of 'on' and 'in' using the boxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 169.0], "base_se_norm": [-0.07311320304870605, 0.1957547664642334], "gt_se_raw": [151.0, 158.0], "gt_se_norm": [0.06839621067047119, 0.11792457103729248], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006676.pt"}
{"idx": 6678, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/--YU8YcWeUU.mp4", "duration": 228.09453333333335, "prompt_text": "Please find the visual event described by the sentence 'When is a man in a blue shirt running on a road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man in a blue shirt running on a road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9824634194374084], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.9912316799163818], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006678.pt"}
{"idx": 6680, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/--YU8YcWeUU.mp4", "duration": 228.09453333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the text 'Tip #2: DON'T Heelstrike!'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the text 'Tip #2: DON'T Heelstrike!'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 116.0], "base_se_norm": [-0.017951011657714844, 0.01712214946746826], "gt_se_raw": [112.0, 116.0], "gt_se_norm": [-0.017951011657714844, 0.01712214946746826], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006680.pt"}
{"idx": 6682, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/--YU8YcWeUU.mp4", "duration": 228.09453333333335, "prompt_text": "Please find the visual event described by the sentence 'When do the two runners introduce themselves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two runners introduce themselves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 24.0], "base_se_norm": [-0.8947804570198059, -0.7895609140396118], "gt_se_raw": [12.0, 19.0], "gt_se_norm": [-0.8947804570198059, -0.8334023952484131], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006682.pt"}
{"idx": 6684, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FWlHYefkQms.mp4", "duration": 235.76793333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time the older man in the black coat points a gun at the others.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time the older man in the black coat points a gun at the others.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 23.0], "base_se_norm": [-0.949102520942688, -0.804892897605896], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.949102520942688, -0.9236537218093872], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006684.pt"}
{"idx": 6686, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FWlHYefkQms.mp4", "duration": 235.76793333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man in the brown coat says 'You were my only friend'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man in the brown coat says 'You were my only friend'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 96.0], "base_se_norm": [-0.2365373969078064, -0.1856398582458496], "gt_se_raw": [106.0, 108.0], "gt_se_norm": [-0.100810706615448, -0.08384484052658081], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006686.pt"}
{"idx": 6688, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wu-gd9tLpmU.mp4", "duration": 221.98843333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the girl counting with the light for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl counting with the light for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 34.0], "base_se_norm": [-0.7927820086479187, -0.6936777234077454], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [-0.7927820086479187, -0.7387251257896423], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006688.pt"}
{"idx": 6690, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wu-gd9tLpmU.mp4", "duration": 221.98843333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me when the girl successfully counts to five in her head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the girl successfully counts to five in her head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [-0.5134881734848022, -0.4594312906265259], "gt_se_raw": [53.0, 60.0], "gt_se_norm": [-0.5224976539611816, -0.4594312906265259], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006690.pt"}
{"idx": 6692, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/reZpm2zH3UQ.mp4", "duration": 282.53333333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the slide about Closed Captioning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the slide about Closed Captioning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 44.0], "base_se_norm": [-0.7380839586257935, -0.6885323524475098], "gt_se_raw": [38.0, 44.0], "gt_se_norm": [-0.7310051918029785, -0.6885323524475098], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006692.pt"}
{"idx": 6694, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/reZpm2zH3UQ.mp4", "duration": 282.53333333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point is the definition of 'Audio Description' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the definition of 'Audio Description' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 151.0], "base_se_norm": [-0.01604527235031128, 0.068900465965271], "gt_se_raw": [139.0, 151.0], "gt_se_norm": [-0.01604527235031128, 0.068900465965271], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006694.pt"}
{"idx": 6696, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/reZpm2zH3UQ.mp4", "duration": 282.53333333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the sources are listed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 279.0 - 282.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the sources are listed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 279.0 - 282.0 seconds.<|im_end|>\n", "base_se_raw": [279.0, 282.0], "base_se_norm": [0.9749882221221924, 0.9962246417999268], "gt_se_raw": [280.0, 282.0], "gt_se_norm": [0.9820671081542969, 0.9962246417999268], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006696.pt"}
{"idx": 6698, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GqSG43pgvZE.mp4", "duration": 223.89033333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a person pouring pink sparkling wine into a glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person pouring pink sparkling wine into a glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 185.0], "base_se_norm": [0.6257959604263306, 0.652594804763794], "gt_se_raw": [183.0, 186.0], "gt_se_norm": [0.6347289085388184, 0.6615277528762817], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006698.pt"}
{"idx": 6700, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GqSG43pgvZE.mp4", "duration": 223.89033333333333, "prompt_text": "Please find the visual event described by the sentence 'When are two women toasting with teacups?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 171.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women toasting with teacups?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 171.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 171.0], "base_se_norm": [0.49180173873901367, 0.5275335311889648], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [-0.4908221364021301, -0.47295624017715454], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006700.pt"}
{"idx": 6702, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5dMmKehyVUs.mp4", "duration": 298.56493333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when a black dog is lying on the carpet in the bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a black dog is lying on the carpet in the bedroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 39.0], "base_se_norm": [-0.879423201084137, -0.7387503385543823], "gt_se_raw": [18.0, 30.0], "gt_se_norm": [-0.879423201084137, -0.7990387082099915], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006702.pt"}
{"idx": 6704, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5dMmKehyVUs.mp4", "duration": 298.56493333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where people are walking down a hallway in a stadium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where people are walking down a hallway in a stadium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 98.0], "base_se_norm": [-0.3636225461959839, -0.3435264229774475], "gt_se_raw": [95.0, 98.0], "gt_se_norm": [-0.3636225461959839, -0.3435264229774475], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006704.pt"}
{"idx": 6706, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5dMmKehyVUs.mp4", "duration": 298.56493333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman come home and greet her dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 178.0 - 218.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman come home and greet her dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 178.0 - 218.0 seconds.<|im_end|>\n", "base_se_raw": [178.0, 218.0], "base_se_norm": [0.19237041473388672, 0.46031880378723145], "gt_se_raw": [179.0, 199.0], "gt_se_norm": [0.19906914234161377, 0.33304333686828613], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006706.pt"}
{"idx": 6708, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QGGyYVsEGGo.mp4", "duration": 240.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when Randy Moss makes a one-handed catch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when Randy Moss makes a one-handed catch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 130.0], "base_se_norm": [-0.0011098384857177734, 0.082131028175354], "gt_se_raw": [120.0, 122.0], "gt_se_norm": [-0.0011098384857177734, 0.015538334846496582], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006708.pt"}
{"idx": 6710, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QGGyYVsEGGo.mp4", "duration": 240.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does a player in a green jersey catch the ball while lying on his back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 199.0 - 204.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a player in a green jersey catch the ball while lying on his back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 199.0 - 204.0 seconds.<|im_end|>\n", "base_se_raw": [199.0, 204.0], "base_se_norm": [0.6564928293228149, 0.698113203048706], "gt_se_raw": [189.0, 192.0], "gt_se_norm": [0.5732519626617432, 0.5982241630554199], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006710.pt"}
{"idx": 6712, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U11DgbefmQQ.mp4", "duration": 310.0, "prompt_text": "Please find the visual event described by the sentence 'When does the mantis shrimp attack the clear shrimp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the mantis shrimp attack the clear shrimp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 158.0], "base_se_norm": [-0.051612913608551025, 0.019354820251464844], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [-0.4516128897666931, -0.43870967626571655], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006712.pt"}
{"idx": 6714, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U11DgbefmQQ.mp4", "duration": 310.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the shrimp carry the crab away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the shrimp carry the crab away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 158.0], "base_se_norm": [-0.051612913608551025, 0.019354820251464844], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.7483870983123779, -0.7225806713104248], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006714.pt"}
{"idx": 6716, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U11DgbefmQQ.mp4", "duration": 310.0, "prompt_text": "Please find the visual event described by the sentence 'When does the mantis shrimp first strike the blue crab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the mantis shrimp first strike the blue crab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 43.0], "base_se_norm": [-0.8774193525314331, -0.7225806713104248], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.6838709712028503, -0.670967698097229], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006716.pt"}
{"idx": 6718, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JaU55dchC-g.mp4", "duration": 272.272, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a man wearing a 'TEAM BOB' cap appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a man wearing a 'TEAM BOB' cap appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [-0.5739554166793823, -0.5372275114059448], "gt_se_raw": [58.0, 63.0], "gt_se_norm": [-0.5739554166793823, -0.5372275114059448], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006718.pt"}
{"idx": 6720, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JaU55dchC-g.mp4", "duration": 272.272, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a woman in a nursing uniform walking towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 152.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a woman in a nursing uniform walking towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 152.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 152.0], "base_se_norm": [0.050420165061950684, 0.11653053760528564], "gt_se_raw": [143.0, 152.0], "gt_se_norm": [0.050420165061950684, 0.11653053760528564], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006720.pt"}
{"idx": 6722, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/14GZruL1hrE.mp4", "duration": 299.2572916666666, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in pink pajamas hide under the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in pink pajamas hide under the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.5388583540916443, -0.5121254920959473], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [-0.5254919528961182, -0.5121254920959473], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006722.pt"}
{"idx": 6724, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/14GZruL1hrE.mp4", "duration": 299.2572916666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man angrily sweeps things off his desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man angrily sweeps things off his desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 111.0], "base_se_norm": [-0.2715298533439636, -0.2581634521484375], "gt_se_raw": [109.0, 111.0], "gt_se_norm": [-0.2715298533439636, -0.2581634521484375], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006724.pt"}
{"idx": 6726, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/14GZruL1hrE.mp4", "duration": 299.2572916666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the police are shining flashlights at the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 262.0 - 265.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the police are shining flashlights at the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 262.0 - 265.0 seconds.<|im_end|>\n", "base_se_raw": [262.0, 265.0], "base_se_norm": [0.7510015964508057, 0.771051287651062], "gt_se_raw": [263.0, 266.0], "gt_se_norm": [0.7576848268508911, 0.7777343988418579], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006726.pt"}
{"idx": 6728, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4eYAVZkCjvY.mp4", "duration": 290.55693333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the little girl dressed as a working mom enter the stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little girl dressed as a working mom enter the stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 81.0], "base_se_norm": [-0.4837501049041748, -0.4424501061439514], "gt_se_raw": [75.0, 81.0], "gt_se_norm": [-0.4837501049041748, -0.4424501061439514], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006728.pt"}
{"idx": 6730, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4eYAVZkCjvY.mp4", "duration": 290.55693333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where Ellen signs the giant check costume.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where Ellen signs the giant check costume.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 137.0], "base_se_norm": [-0.11893349885940552, -0.05698347091674805], "gt_se_raw": [140.0, 146.0], "gt_se_norm": [-0.03633350133895874, 0.0049664974212646484], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006730.pt"}
{"idx": 6732, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XzP3jOJn5c.mp4", "duration": 223.24, "prompt_text": "Please find the visual event described by the sentence 'When is the woman sitting in the armchair and reading the newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman sitting in the armchair and reading the newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.928328275680542, -0.856656551361084], "gt_se_raw": [7.0, 16.0], "gt_se_norm": [-0.9372872114181519, -0.856656551361084], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006732.pt"}
{"idx": 6734, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XzP3jOJn5c.mp4", "duration": 223.24, "prompt_text": "Please find the visual event described by the sentence 'At what point are multiple satellites shown orbiting the Earth in formation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are multiple satellites shown orbiting the Earth in formation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 132.0], "base_se_norm": [0.11987102031707764, 0.18258368968963623], "gt_se_raw": [127.0, 133.0], "gt_se_norm": [0.13778889179229736, 0.19154274463653564], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006734.pt"}
{"idx": 6736, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XzP3jOJn5c.mp4", "duration": 223.24, "prompt_text": "Please find the visual event described by the sentence 'When are the parasol effect and the greenhouse effect depicted as a balancing scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the parasol effect and the greenhouse effect depicted as a balancing scale?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 109.0], "base_se_norm": [-0.0861852765083313, -0.02347254753112793], "gt_se_raw": [102.0, 109.0], "gt_se_norm": [-0.0861852765083313, -0.02347254753112793], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006736.pt"}
{"idx": 6738, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IW40rIEXKXs.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man says 'that US overpriced housing bubble that burst'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man says 'that US overpriced housing bubble that burst'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.9099999070167542, -0.8829998970031738], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.6489996910095215, -0.6219996213912964], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006738.pt"}
{"idx": 6740, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IW40rIEXKXs.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man gestures with both hands up, palms facing forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man gestures with both hands up, palms facing forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 77.0], "base_se_norm": [-0.3339993357658386, -0.3069993257522583], "gt_se_raw": [75.0, 76.0], "gt_se_norm": [-0.3249993324279785, -0.3159993290901184], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006740.pt"}
{"idx": 6742, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bzua8Zvlppo.mp4", "duration": 209.0088, "prompt_text": "Please find the visual event described by the sentence 'When is the speaker Urvashi Vaid shown talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the speaker Urvashi Vaid shown talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 37.0], "base_se_norm": [-0.9330171942710876, -0.6459479331970215], "gt_se_raw": [7.0, 37.0], "gt_se_norm": [-0.9330171942710876, -0.6459479331970215], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006742.pt"}
{"idx": 6744, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bzua8Zvlppo.mp4", "duration": 209.0088, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker mentions the 'foundation dance'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker mentions the 'foundation dance'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 162.0], "base_se_norm": [0.5214669704437256, 0.5501738786697388], "gt_se_raw": [123.0, 125.0], "gt_se_norm": [0.17698395252227783, 0.19612181186676025], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006744.pt"}
{"idx": 6746, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bzua8Zvlppo.mp4", "duration": 209.0088, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker in the 'GOOD OLD DAYS' t-shirt appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 150.0 - 178.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker in the 'GOOD OLD DAYS' t-shirt appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 150.0 - 178.0 seconds.<|im_end|>\n", "base_se_raw": [150.0, 178.0], "base_se_norm": [0.43534624576568604, 0.703277587890625], "gt_se_raw": [150.0, 179.0], "gt_se_norm": [0.43534624576568604, 0.7128465175628662], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006746.pt"}
{"idx": 6748, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a5V6gdu5ih8.mp4", "duration": 213.76, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where a hyena bares its teeth at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where a hyena bares its teeth at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.8222305178642273, -0.7941616773605347], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.8222305178642273, -0.8035179376602173], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006748.pt"}
{"idx": 6750, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a5V6gdu5ih8.mp4", "duration": 213.76, "prompt_text": "Please find the visual event described by the sentence 'At what point does the lion sit down while being surrounded by the hyenas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the lion sit down while being surrounded by the hyenas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 68.0], "base_se_norm": [-0.5134730339050293, -0.363772451877594], "gt_se_raw": [51.0, 68.0], "gt_se_norm": [-0.5228292942047119, -0.363772451877594], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006750.pt"}
{"idx": 6752, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tndpRiDf9Z0.mp4", "duration": 215.840625, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cube's interior light up with multiple red dots?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cube's interior light up with multiple red dots?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 25.0], "base_se_norm": [-0.879540741443634, -0.7683476209640503], "gt_se_raw": [13.0, 25.0], "gt_se_norm": [-0.879540741443634, -0.7683476209640503], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006752.pt"}
{"idx": 6754, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tndpRiDf9Z0.mp4", "duration": 215.840625, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the man in the red vest introduces himself as 'Crothers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the man in the red vest introduces himself as 'Crothers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 62.0], "base_se_norm": [-0.5366951823234558, -0.4255020022392273], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.5181629657745361, -0.49963080883026123], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006754.pt"}
{"idx": 6756, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tndpRiDf9Z0.mp4", "duration": 215.840625, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman wakes up in bed, looking confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman wakes up in bed, looking confused.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 168.0], "base_se_norm": [0.4825754165649414, 0.5567041635513306], "gt_se_raw": [160.0, 167.0], "gt_se_norm": [0.4825754165649414, 0.5474381446838379], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006756.pt"}
{"idx": 6758, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t1PGWO2Lvmw.mp4", "duration": 203.66179166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the process where the cream cheese filling is beaten with a hand mixer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process where the cream cheese filling is beaten with a hand mixer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 55.0], "base_se_norm": [-0.5188100934028625, -0.45988887548446655], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [-0.5188100934028625, -0.45988887548446655], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006758.pt"}
{"idx": 6760, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t1PGWO2Lvmw.mp4", "duration": 203.66179166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where sliced strawberries are placed on top of the cream-spread crepe strips.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where sliced strawberries are placed on top of the cream-spread crepe strips.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 103.0], "base_se_norm": [-0.04744035005569458, 0.01148080825805664], "gt_se_raw": [97.0, 103.0], "gt_se_norm": [-0.04744035005569458, 0.01148080825805664], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006760.pt"}
{"idx": 6762, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7FgIp5QrZik.mp4", "duration": 301.926625, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the blue chameleon in the pond, the yellow chameleon on the sand, and the green chameleon on the leaf all on one page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the blue chameleon in the pond, the yellow chameleon on the sand, and the green chameleon on the leaf all on one page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.7880280017852783, -0.7350349426269531], "gt_se_raw": [23.0, 39.0], "gt_se_norm": [-0.847645103931427, -0.7416591048240662], "start_token_span": [87, 91], "end_token_span": [93, 97], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006762.pt"}
{"idx": 6764, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7FgIp5QrZik.mp4", "duration": 301.926625, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Leon becomes orange while in the blue pond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Leon becomes orange while in the blue pond.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 84.0], "base_se_norm": [-0.5826801061630249, -0.4435734152793884], "gt_se_raw": [67.0, 72.0], "gt_se_norm": [-0.5561835765838623, -0.5230629444122314], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006764.pt"}
{"idx": 6766, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7FgIp5QrZik.mp4", "duration": 301.926625, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where the chameleon's parents, colored red, are searching for the lost chameleons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 225.0 - 251.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where the chameleon's parents, colored red, are searching for the lost chameleons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 225.0 - 251.0 seconds.<|im_end|>\n", "base_se_raw": [225.0, 251.0], "base_se_norm": [0.4904283285140991, 0.6626555919647217], "gt_se_raw": [226.0, 234.0], "gt_se_norm": [0.4970524311065674, 0.550045371055603], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006766.pt"}
{"idx": 6768, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwXx63AoSwU.mp4", "duration": 277.4583333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman with black and white hair is singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman with black and white hair is singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 113.0], "base_se_norm": [-0.2503378987312317, -0.1854633092880249], "gt_se_raw": [105.0, 113.0], "gt_se_norm": [-0.24312961101531982, -0.1854633092880249], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006768.pt"}
{"idx": 6770, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwXx63AoSwU.mp4", "duration": 277.4583333333333, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a man dressed in purple with a skull on his hat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a man dressed in purple with a skull on his hat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 128.0], "base_se_norm": [-0.17825502157211304, -0.07733899354934692], "gt_se_raw": [124.0, 126.0], "gt_se_norm": [-0.10617214441299438, -0.09175556898117065], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006770.pt"}
{"idx": 6772, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y9PV2rfjJo4.mp4", "duration": 236.5363, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention the social moments she has in the morning at work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention the social moments she has in the morning at work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 154.0], "base_se_norm": [0.24293816089630127, 0.3021256923675537], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [-0.8647142052650452, -0.8139820098876953], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006772.pt"}
{"idx": 6774, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y9PV2rfjJo4.mp4", "duration": 236.5363, "prompt_text": "Please find the visual event described by the sentence 'Show me when the young man, the young woman, and a black dog are sitting together by the lockers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the young man, the young woman, and a black dog are sitting together by the lockers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 146.0], "base_se_norm": [0.1414738893508911, 0.2344827651977539], "gt_se_raw": [135.0, 146.0], "gt_se_norm": [0.1414738893508911, 0.2344827651977539], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006774.pt"}
{"idx": 6776, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y9PV2rfjJo4.mp4", "duration": 236.5363, "prompt_text": "Please find the visual event described by the sentence 'When do we see a young man interacting with a group of young women in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see a young man interacting with a group of young women in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 146.0], "base_se_norm": [0.1414738893508911, 0.2344827651977539], "gt_se_raw": [175.0, 184.0], "gt_se_norm": [0.47968828678131104, 0.5557866096496582], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006776.pt"}
{"idx": 6778, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jofNR_WkoCE.mp4", "duration": 224.8, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the fox costume sing while sitting on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the fox costume sing while sitting on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [-0.6530249118804932, -0.6352313160896301], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.6530249118804932, -0.6352313160896301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006778.pt"}
{"idx": 6780, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jofNR_WkoCE.mp4", "duration": 224.8, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a group of people is dancing in a dark forest with glowing green lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a group of people is dancing in a dark forest with glowing green lights.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 134.0], "base_se_norm": [0.023131608963012695, 0.1921708583831787], "gt_se_raw": [116.0, 122.0], "gt_se_norm": [0.0320284366607666, 0.08540928363800049], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006780.pt"}
{"idx": 6782, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLjHyP683QU.mp4", "duration": 205.04, "prompt_text": "Please find the visual event described by the sentence 'When can we see the sun rising over the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the sun rising over the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [-0.7561451196670532, -0.6586031913757324], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.7463909387588501, -0.7073741555213928], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006782.pt"}
{"idx": 6784, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLjHyP683QU.mp4", "duration": 205.04, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the starfish is unwrapped from the star-shaped gift.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the starfish is unwrapped from the star-shaped gift.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 111.0], "base_se_norm": [0.024190425872802734, 0.08271563053131104], "gt_se_raw": [105.0, 109.0], "gt_se_norm": [0.024190425872802734, 0.06320726871490479], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006784.pt"}
{"idx": 6786, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLjHyP683QU.mp4", "duration": 205.04, "prompt_text": "Please find the visual event described by the sentence 'When does the little fish blow out the candles on the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 186.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little fish blow out the candles on the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 186.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 186.0], "base_se_norm": [0.7655092477798462, 0.8142801523208618], "gt_se_raw": [182.0, 185.0], "gt_se_norm": [0.7752634286880493, 0.8045259714126587], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006786.pt"}
{"idx": 6788, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zRgZ3sWvnqs.mp4", "duration": 269.8112083333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the animation showing a document with code connecting to a computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the animation showing a document with code connecting to a computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 61.0], "base_se_norm": [-0.5774823427200317, -0.5478320121765137], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [-0.5774823427200317, -0.5552445650100708], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006788.pt"}
{"idx": 6790, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zRgZ3sWvnqs.mp4", "duration": 269.8112083333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the speaker mentions 'ONIX 3.0'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 201.0 - 205.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the speaker mentions 'ONIX 3.0'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 201.0 - 205.0 seconds.<|im_end|>\n", "base_se_raw": [201.0, 205.0], "base_se_norm": [0.4899306297302246, 0.5195809602737427], "gt_se_raw": [199.0, 204.0], "gt_se_norm": [0.4751054048538208, 0.5121684074401855], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006790.pt"}
{"idx": 6792, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hPLgfGX1I5Y.mp4", "duration": 206.8733333333333, "prompt_text": "Please find the visual event described by the sentence 'When do we see the evolution of man depicted as walking in a line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see the evolution of man depicted as walking in a line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.6422932147979736, -0.5842866897583008], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [-0.6422932147979736, -0.6132899522781372], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006792.pt"}
{"idx": 6794, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hPLgfGX1I5Y.mp4", "duration": 206.8733333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show various dinosaur fossils?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show various dinosaur fossils?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 72.0], "base_se_norm": [-0.4199349284172058, -0.3039218783378601], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [-0.34259289503097534, -0.3039218783378601], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006794.pt"}
{"idx": 6796, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hPLgfGX1I5Y.mp4", "duration": 206.8733333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the animation illustrating how a small dinosaur might use its feathers to help it run up an incline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the animation illustrating how a small dinosaur might use its feathers to help it run up an incline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 182.0], "base_se_norm": [0.5661756992340088, 0.759530782699585], "gt_se_raw": [162.0, 175.0], "gt_se_norm": [0.5661756992340088, 0.6918565034866333], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006796.pt"}
{"idx": 6798, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c346ms-PaeE.mp4", "duration": 207.33333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the pig with a crown kills the man on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the pig with a crown kills the man on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 188.0], "base_se_norm": [0.7459807395935059, 0.8135048151016235], "gt_se_raw": [186.0, 188.0], "gt_se_norm": [0.7942122220993042, 0.8135048151016235], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006798.pt"}
{"idx": 6800, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c346ms-PaeE.mp4", "duration": 207.33333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man in the striped suit is killed by an arrow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man in the striped suit is killed by an arrow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 182.0], "base_se_norm": [0.7073955535888672, 0.7556270360946655], "gt_se_raw": [142.0, 145.0], "gt_se_norm": [0.3697749376296997, 0.3987138271331787], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006800.pt"}
{"idx": 6802, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1fB0BKm9dTE.mp4", "duration": 311.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the puma is in the sandbox.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the puma is in the sandbox.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, -0.8457252979278564], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9871437549591064], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006802.pt"}
{"idx": 6804, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1fB0BKm9dTE.mp4", "duration": 311.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman is feeding a tiger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman is feeding a tiger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 51.0], "base_se_norm": [-0.7750160694122314, -0.6721662282943726], "gt_se_raw": [38.0, 45.0], "gt_se_norm": [-0.7557317018508911, -0.7107349634170532], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006804.pt"}
{"idx": 6806, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1fB0BKm9dTE.mp4", "duration": 311.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of the three lemurs foraging on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 271.0 - 305.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of the three lemurs foraging on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 271.0 - 305.0 seconds.<|im_end|>\n", "base_se_raw": [271.0, 305.0], "base_se_norm": [0.742018461227417, 0.9605742692947388], "gt_se_raw": [271.0, 296.0], "gt_se_norm": [0.742018461227417, 0.9027212858200073], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006806.pt"}
{"idx": 6808, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TSVj63OE9qM.mp4", "duration": 212.04, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the purple-skinned sea witch's solo performance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the purple-skinned sea witch's solo performance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 78.0], "base_se_norm": [-0.5189586877822876, -0.2642897367477417], "gt_se_raw": [52.0, 82.0], "gt_se_norm": [-0.5095264911651611, -0.22656100988388062], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006808.pt"}
{"idx": 6810, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TSVj63OE9qM.mp4", "duration": 212.04, "prompt_text": "Please find the visual event described by the sentence 'Show me when green flames appear at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when green flames appear at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 163.0], "base_se_norm": [0.30164122581481934, 0.5374457836151123], "gt_se_raw": [138.0, 163.0], "gt_se_norm": [0.30164122581481934, 0.5374457836151123], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006810.pt"}
{"idx": 6812, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OstpJCHpTis.mp4", "duration": 280.0464333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is a man walking down a grand, marbled staircase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is a man walking down a grand, marbled staircase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.9071582555770874, -0.871449887752533], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.9000166058540344, -0.871449887752533], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006812.pt"}
{"idx": 6814, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OstpJCHpTis.mp4", "duration": 280.0464333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part of the animation that illustrates the application of dried smallpox scabs to a scratch on the arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 217.0 - 225.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part of the animation that illustrates the application of dried smallpox scabs to a scratch on the arm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 217.0 - 225.0 seconds.<|im_end|>\n", "base_se_raw": [217.0, 225.0], "base_se_norm": [0.5497429370880127, 0.6068762540817261], "gt_se_raw": [115.0, 121.0], "gt_se_norm": [-0.17870765924453735, -0.13585764169692993], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006814.pt"}
{"idx": 6816, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OstpJCHpTis.mp4", "duration": 280.0464333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a cartoon depicts people growing cow parts as a result of vaccination.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 268.0 - 276.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a cartoon depicts people growing cow parts as a result of vaccination.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 268.0 - 276.0 seconds.<|im_end|>\n", "base_se_raw": [268.0, 276.0], "base_se_norm": [0.9139682054519653, 0.9711016416549683], "gt_se_raw": [272.0, 276.0], "gt_se_norm": [0.9425349235534668, 0.9711016416549683], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006816.pt"}
{"idx": 6818, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ovyEMlNZlTE.mp4", "duration": 242.367125, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when Steve Martin introduces himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when Steve Martin introduces himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 37.0], "base_se_norm": [-0.8019533157348633, -0.6946780681610107], "gt_se_raw": [35.0, 36.0], "gt_se_norm": [-0.7111819386482239, -0.7029299736022949], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006818.pt"}
{"idx": 6820, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ovyEMlNZlTE.mp4", "duration": 242.367125, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the purple jacket point at the painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 188.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the purple jacket point at the painting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 188.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [188.0, 193.0], "base_se_norm": [0.5513654947280884, 0.5926252603530884], "gt_se_raw": [188.0, 193.0], "gt_se_norm": [0.5513654947280884, 0.5926252603530884], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006820.pt"}
{"idx": 6822, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZKVBUjaxOs.mp4", "duration": 251.5513, "prompt_text": "Please find the visual event described by the sentence 'When are the sweet potato slices being arranged on the baking trays?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the sweet potato slices being arranged on the baking trays?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.8250854015350342, -0.777381420135498], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.8171347379684448, -0.777381420135498], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006822.pt"}
{"idx": 6824, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZKVBUjaxOs.mp4", "duration": 251.5513, "prompt_text": "Please find the visual event described by the sentence 'At what point are the shrimp and tomatoes added to the baking pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the shrimp and tomatoes added to the baking pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 148.0], "base_se_norm": [0.11309301853179932, 0.17669832706451416], "gt_se_raw": [141.0, 144.0], "gt_se_norm": [0.12104368209838867, 0.14489567279815674], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006824.pt"}
{"idx": 6826, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZKVBUjaxOs.mp4", "duration": 251.5513, "prompt_text": "Please find the visual event described by the sentence 'When is someone dipping a piece of toast into the Tuscan white beans?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is someone dipping a piece of toast into the Tuscan white beans?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 113.0], "base_se_norm": [-0.12542688846588135, -0.10157489776611328], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [-0.12542688846588135, -0.10952556133270264], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006826.pt"}
{"idx": 6828, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4ztYjFxwmI.mp4", "duration": 280.34673333333336, "prompt_text": "Please find the visual event described by the sentence 'Show the part where salt is being added to the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where salt is being added to the ice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 68.0], "base_se_norm": [-0.6361648440361023, -0.5148864984512329], "gt_se_raw": [49.0, 68.0], "gt_se_norm": [-0.6504328846931458, -0.5148864984512329], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006828.pt"}
{"idx": 6830, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4ztYjFxwmI.mp4", "duration": 280.34673333333336, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the small bag is placed inside the larger bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the small bag is placed inside the larger bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 159.0], "base_se_norm": [-0.12251520156860352, 0.13430964946746826], "gt_se_raw": [173.0, 184.0], "gt_se_norm": [0.23418593406677246, 0.31266021728515625], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006830.pt"}
{"idx": 6832, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-GRl8urBK4c.mp4", "duration": 302.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the hands rolling the red wool tightly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the hands rolling the red wool tightly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 30.0], "base_se_norm": [-0.8742415904998779, -0.8014340996742249], "gt_se_raw": [19.0, 28.0], "gt_se_norm": [-0.8742415904998779, -0.8146718144416809], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006832.pt"}
{"idx": 6834, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-GRl8urBK4c.mp4", "duration": 302.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the white dots are being added to the mushroom cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the white dots are being added to the mushroom cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 159.0], "base_se_norm": [-0.1991174817085266, 0.052399396896362305], "gt_se_raw": [121.0, 159.0], "gt_se_norm": [-0.1991174817085266, 0.052399396896362305], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006834.pt"}
{"idx": 6836, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-GRl8urBK4c.mp4", "duration": 302.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two parts of the mushroom attached together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 254.0 - 278.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two parts of the mushroom attached together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 254.0 - 278.0 seconds.<|im_end|>\n", "base_se_raw": [254.0, 278.0], "base_se_norm": [0.6811914443969727, 0.8400441408157349], "gt_se_raw": [255.0, 262.0], "gt_se_norm": [0.6878103017807007, 0.7341424226760864], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006836.pt"}
{"idx": 6838, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PB9JxYt2ls.mp4", "duration": 239.3391, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the paper is folded diagonally to the center crease.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the paper is folded diagonally to the center crease.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 66.0], "base_se_norm": [-0.6824588775634766, -0.4484812617301941], "gt_se_raw": [39.0, 67.0], "gt_se_norm": [-0.6741025447845459, -0.4401249289512634], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006838.pt"}
{"idx": 6840, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PB9JxYt2ls.mp4", "duration": 239.3391, "prompt_text": "Please find the visual event described by the sentence 'At what time are the top, bottom, and sides folded back to shape the pumpkin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 191.0 - 211.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the top, bottom, and sides folded back to shape the pumpkin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 191.0 - 211.0 seconds.<|im_end|>\n", "base_se_raw": [191.0, 211.0], "base_se_norm": [0.5960618257522583, 0.7631887197494507], "gt_se_raw": [191.0, 210.0], "gt_se_norm": [0.5960618257522583, 0.75483238697052], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006840.pt"}
{"idx": 6842, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5H8zBb3iao.mp4", "duration": 226.97674999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone doing a one-handed handstand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone doing a one-handed handstand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.8237705230712891, -0.7973361015319824], "gt_se_raw": [20.0, 21.0], "gt_se_norm": [-0.8237705230712891, -0.8149590492248535], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006842.pt"}
{"idx": 6844, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5H8zBb3iao.mp4", "duration": 226.97674999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man in the dinosaur costume dancing on the ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man in the dinosaur costume dancing on the ice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.7444672584533691, -0.7004098296165466], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.3920081853866577, -0.3743852376937866], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006844.pt"}
{"idx": 6846, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5H8zBb3iao.mp4", "duration": 226.97674999999998, "prompt_text": "Please find the visual event described by the sentence 'When do all the players huddle together and put their hands in the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the players huddle together and put their hands in the center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 87.0], "base_se_norm": [-0.26864755153656006, -0.23340165615081787], "gt_se_raw": [205.0, 207.0], "gt_se_norm": [0.8063524961471558, 0.8239754438400269], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006846.pt"}
{"idx": 6848, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Yrcr9QOnqB4.mp4", "duration": 313.4798333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man attacks the projector with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man attacks the projector with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 86.0], "base_se_norm": [-0.5406402945518494, -0.4513203501701355], "gt_se_raw": [82.0, 86.0], "gt_se_norm": [-0.4768403172492981, -0.4513203501701355], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006848.pt"}
{"idx": 6850, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Yrcr9QOnqB4.mp4", "duration": 313.4798333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man use a lighter to set the film reels on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man use a lighter to set the film reels on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 138.0], "base_se_norm": [-0.1961205005645752, -0.1195605993270874], "gt_se_raw": [127.0, 138.0], "gt_se_norm": [-0.18974053859710693, -0.1195605993270874], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006850.pt"}
{"idx": 6852, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ml9EdqgtVfU.mp4", "duration": 299.23333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the process of selecting a picture from the local computer to upload.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of selecting a picture from the local computer to upload.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 50.0], "base_se_norm": [-0.93316251039505, -0.665812611579895], "gt_se_raw": [20.0, 25.0], "gt_se_norm": [-0.8663250207901001, -0.8329063057899475], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006852.pt"}
{"idx": 6854, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ml9EdqgtVfU.mp4", "duration": 299.23333333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the image's alignment is adjusted to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the image's alignment is adjusted to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 88.0], "base_se_norm": [-0.4385652542114258, -0.41183024644851685], "gt_se_raw": [113.0, 117.0], "gt_se_norm": [-0.24473655223846436, -0.21800154447555542], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006854.pt"}
{"idx": 6856, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ml9EdqgtVfU.mp4", "duration": 299.23333333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me when the size of the first embedded video is changed to 'Large'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 263.0 - 268.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the size of the first embedded video is changed to 'Large'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 263.0 - 268.0 seconds.<|im_end|>\n", "base_se_raw": [263.0, 268.0], "base_se_norm": [0.757825493812561, 0.7912442684173584], "gt_se_raw": [264.0, 270.0], "gt_se_norm": [0.7645093202590942, 0.8046118021011353], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006856.pt"}
{"idx": 6858, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-zXT5bIBM0.mp4", "duration": 201.73486666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is putting a glove on a little girl's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is putting a glove on a little girl's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.5538698434829712, -0.48447185754776], "gt_se_raw": [46.0, 47.0], "gt_se_norm": [-0.5439558625221252, -0.5340418815612793], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006858.pt"}
{"idx": 6860, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-zXT5bIBM0.mp4", "duration": 201.73486666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the girl lie on the ground next to a grandfather clock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl lie on the ground next to a grandfather clock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 76.0], "base_se_norm": [-0.2961058020591736, -0.2465357780456543], "gt_se_raw": [72.0, 76.0], "gt_se_norm": [-0.28619182109832764, -0.2465357780456543], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006860.pt"}
{"idx": 6862, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bv2vT665bGI.mp4", "duration": 218.0, "prompt_text": "Please find the visual event described by the sentence 'When does the presenter explain the materials needed for the experiment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the presenter explain the materials needed for the experiment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 49.0], "base_se_norm": [-0.8256880640983582, -0.5504587292671204], "gt_se_raw": [20.0, 33.0], "gt_se_norm": [-0.8165137767791748, -0.6972477436065674], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006862.pt"}
{"idx": 6864, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bv2vT665bGI.mp4", "duration": 218.0, "prompt_text": "Please find the visual event described by the sentence 'When does the person insert the copper clips into the lemon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person insert the copper clips into the lemon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 66.0], "base_se_norm": [-0.541284441947937, -0.39449542760849], "gt_se_raw": [51.0, 60.0], "gt_se_norm": [-0.5321100950241089, -0.44954127073287964], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006864.pt"}
{"idx": 6866, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bv2vT665bGI.mp4", "duration": 218.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the fire successfully started using the lemon battery?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the fire successfully started using the lemon battery?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 158.0], "base_se_norm": [0.3486238718032837, 0.4495413303375244], "gt_se_raw": [149.0, 155.0], "gt_se_norm": [0.3669724464416504, 0.42201828956604004], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006866.pt"}
{"idx": 6868, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dxvFck5LBkU.mp4", "duration": 291.16, "prompt_text": "Please find the visual event described by the sentence 'Show the process of mixing all the ingredients in a pink bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of mixing all the ingredients in a pink bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 104.0], "base_se_norm": [-0.37491416931152344, -0.2856161594390869], "gt_se_raw": [91.0, 103.0], "gt_se_norm": [-0.37491416931152344, -0.29248523712158203], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006868.pt"}
{"idx": 6870, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dxvFck5LBkU.mp4", "duration": 291.16, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman takes the cooked cookies out of the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman takes the cooked cookies out of the oven.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 173.0], "base_se_norm": [0.06470668315887451, 0.18835008144378662], "gt_se_raw": [168.0, 173.0], "gt_se_norm": [0.15400469303131104, 0.18835008144378662], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006870.pt"}
{"idx": 6872, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zNKWz8hRil4.mp4", "duration": 260.6770833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the children seen getting off the blue bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children seen getting off the blue bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.9079320430755615, -0.8618981242179871], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.9079320430755615, -0.892587423324585], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006872.pt"}
{"idx": 6874, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zNKWz8hRil4.mp4", "duration": 260.6770833333333, "prompt_text": "Please find the visual event described by the sentence 'When do we see people using braille typewriters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see people using braille typewriters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 32.0], "base_se_norm": [-0.8618981242179871, -0.7544854879379272], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [-0.47828173637390137, -0.44759243726730347], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006874.pt"}
{"idx": 6876, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zNKWz8hRil4.mp4", "duration": 260.6770833333333, "prompt_text": "Please find the visual event described by the sentence 'When do the dancers in orange straw skirts perform on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 148.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the dancers in orange straw skirts perform on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 148.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [148.0, 155.0], "base_se_norm": [0.13550448417663574, 0.18921077251434326], "gt_se_raw": [147.0, 154.0], "gt_se_norm": [0.12783217430114746, 0.18153846263885498], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006876.pt"}
{"idx": 6878, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kW1e1qTG_p4.mp4", "duration": 229.3291, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a person is running on a treadmill with a breathing apparatus on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 206.0 - 209.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a person is running on a treadmill with a breathing apparatus on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 206.0 - 209.0 seconds.<|im_end|>\n", "base_se_raw": [206.0, 209.0], "base_se_norm": [0.7965447902679443, 0.822708010673523], "gt_se_raw": [206.0, 209.0], "gt_se_norm": [0.7965447902679443, 0.822708010673523], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006878.pt"}
{"idx": 6880, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kW1e1qTG_p4.mp4", "duration": 229.3291, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of an ultrasound probe being applied to someone's skin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of an ultrasound probe being applied to someone's skin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 146.0], "base_se_norm": [0.23839497566223145, 0.2732793092727661], "gt_se_raw": [142.0, 144.0], "gt_se_norm": [0.23839497566223145, 0.255837082862854], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006880.pt"}
{"idx": 6882, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mVLrBJYGxk4.mp4", "duration": 303.55325, "prompt_text": "Please find the visual event described by the sentence 'When can we see the family having dinner without their masks on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the family having dinner without their masks on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.6639798879623413, -0.6376253366470337], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.6639798879623413, -0.6310367584228516], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006882.pt"}
{"idx": 6884, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mVLrBJYGxk4.mp4", "duration": 303.55325, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the grown-up son leaves his parents and meets his future wife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the grown-up son leaves his parents and meets his future wife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 133.0], "base_se_norm": [-0.16324400901794434, -0.12371224164962769], "gt_se_raw": [128.0, 133.0], "gt_se_norm": [-0.15665537118911743, -0.12371224164962769], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006884.pt"}
{"idx": 6886, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mVLrBJYGxk4.mp4", "duration": 303.55325, "prompt_text": "Please find the visual event described by the sentence 'At what point does the old man take off his mask for the last time before his end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 219.0 - 224.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the old man take off his mask for the last time before his end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 219.0 - 224.0 seconds.<|im_end|>\n", "base_se_raw": [219.0, 224.0], "base_se_norm": [0.44290995597839355, 0.4758530855178833], "gt_se_raw": [219.0, 222.0], "gt_se_norm": [0.44290995597839355, 0.4626758098602295], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006886.pt"}
{"idx": 6888, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcIQiI_Yfg4.mp4", "duration": 250.25, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when lava flows into the ocean, as seen from an underwater perspective.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 213.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when lava flows into the ocean, as seen from an underwater perspective.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 213.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 213.0], "base_se_norm": [0.6303696632385254, 0.7022976875305176], "gt_se_raw": [204.0, 213.0], "gt_se_norm": [0.6303696632385254, 0.7022976875305176], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006888.pt"}
{"idx": 6890, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcIQiI_Yfg4.mp4", "duration": 250.25, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see an aerial view of a crater emitting a massive plume of smoke against a blue sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see an aerial view of a crater emitting a massive plume of smoke against a blue sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 56.0], "base_se_norm": [-0.6563436985015869, -0.5524475574493408], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.6563436985015869, -0.6163836121559143], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006890.pt"}
{"idx": 6892, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wBU9N35ZHIw.mp4", "duration": 308.12, "prompt_text": "Please find the visual event described by the sentence 'When does the person cut a triangular slice of cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person cut a triangular slice of cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 151.0], "base_se_norm": [-0.16266387701034546, -0.019862353801727295], "gt_se_raw": [13.0, 39.0], "gt_se_norm": [-0.9156172871589661, -0.7468518614768982], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006892.pt"}
{"idx": 6894, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wBU9N35ZHIw.mp4", "duration": 308.12, "prompt_text": "Please find the visual event described by the sentence 'Can you specify the time range when a copy of the 'Nature' magazine is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you specify the time range when a copy of the 'Nature' magazine is displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 114.0], "base_se_norm": [-0.4287939667701721, -0.26002854108810425], "gt_se_raw": [88.0, 98.0], "gt_se_norm": [-0.4287939667701721, -0.36388421058654785], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006894.pt"}
{"idx": 6896, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wBU9N35ZHIw.mp4", "duration": 308.12, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a rubber band is used to hold the cake together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 172.0 - 176.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a rubber band is used to hold the cake together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 172.0 - 176.0 seconds.<|im_end|>\n", "base_se_raw": [172.0, 176.0], "base_se_norm": [0.11644816398620605, 0.14241206645965576], "gt_se_raw": [171.0, 175.0], "gt_se_norm": [0.10995721817016602, 0.13592112064361572], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006896.pt"}
{"idx": 6898, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihCciBFdbkw.mp4", "duration": 286.1859, "prompt_text": "Please find the visual event described by the sentence 'Show the part where there's a rapid-fire montage of characters saying the name 'Hiro'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where there's a rapid-fire montage of characters saying the name 'Hiro'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 92.0], "base_se_norm": [-0.5108075141906738, -0.3570612668991089], "gt_se_raw": [71.0, 82.0], "gt_se_norm": [-0.5038190484046936, -0.42694592475891113], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006898.pt"}
{"idx": 6900, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihCciBFdbkw.mp4", "duration": 286.1859, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the final battle scene is compared to a scene from The Avengers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the final battle scene is compared to a scene from The Avengers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.7763691544532776, -0.7554037570953369], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.7344383597373962, -0.7064844965934753], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006900.pt"}
{"idx": 6902, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgfvmHeOiqQ.mp4", "duration": 245.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the boy's heart first leap out of his chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy's heart first leap out of his chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 60.0], "base_se_norm": [-0.5768774747848511, -0.5117816925048828], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [-0.5768774747848511, -0.5280556082725525], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006902.pt"}
{"idx": 6904, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgfvmHeOiqQ.mp4", "duration": 245.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the two boys kiss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the two boys kiss.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 89.0], "base_se_norm": [-0.3083573579788208, -0.27580946683883667], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [-0.3083573579788208, -0.2839464545249939], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006904.pt"}
{"idx": 6906, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgfvmHeOiqQ.mp4", "duration": 245.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'When do the boys' hearts glow together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 224.0 - 227.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the boys' hearts glow together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 224.0 - 227.0 seconds.<|im_end|>\n", "base_se_raw": [224.0, 227.0], "base_se_norm": [0.8226817846298218, 0.8470927476882935], "gt_se_raw": [225.0, 227.0], "gt_se_norm": [0.830818772315979, 0.8470927476882935], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006906.pt"}
{"idx": 6908, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2JduAOOpOSI.mp4", "duration": 235.81891666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the long-haired man enter and sit at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the long-haired man enter and sit at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 52.0], "base_se_norm": [-0.6268323659896851, -0.5589836835861206], "gt_se_raw": [44.0, 49.0], "gt_se_norm": [-0.6268323659896851, -0.5844268798828125], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006908.pt"}
{"idx": 6910, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2JduAOOpOSI.mp4", "duration": 235.81891666666664, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boss hands a pink and blue retro phone to the bald man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boss hands a pink and blue retro phone to the bald man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 156.0], "base_se_norm": [0.20431387424468994, 0.32304906845092773], "gt_se_raw": [141.0, 144.0], "gt_se_norm": [0.1958327293395996, 0.22127604484558105], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006910.pt"}
{"idx": 6912, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K6M67cACkcQ.mp4", "duration": 321.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the slide showing the steps for 'Special Access in Dropbox' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the slide showing the steps for 'Special Access in Dropbox' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 43.0], "base_se_norm": [-0.9128450155258179, -0.7323095798492432], "gt_se_raw": [14.0, 42.0], "gt_se_norm": [-0.9128450155258179, -0.7385349273681641], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006912.pt"}
{"idx": 6914, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K6M67cACkcQ.mp4", "duration": 321.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the presenter choose a student named 'Harry Potter' for special quiz access?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 279.0 - 285.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the presenter choose a student named 'Harry Potter' for special quiz access?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 279.0 - 285.0 seconds.<|im_end|>\n", "base_se_raw": [279.0, 285.0], "base_se_norm": [0.7368749380111694, 0.7742270231246948], "gt_se_raw": [280.0, 284.0], "gt_se_norm": [0.7431002855300903, 0.7680016756057739], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006914.pt"}
{"idx": 6916, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K6M67cACkcQ.mp4", "duration": 321.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the presenter click the 'Save and Close' button after setting up special access for the research paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the presenter click the 'Save and Close' button after setting up special access for the research paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 151.0], "base_se_norm": [-0.08487236499786377, -0.0599709153175354], "gt_se_raw": [147.0, 151.0], "gt_se_norm": [-0.08487236499786377, -0.0599709153175354], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006916.pt"}
{"idx": 6918, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/13S9n41lme8.mp4", "duration": 224.55766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a close-up of a horn shark's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a close-up of a horn shark's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.7862464189529419, -0.7328080534934998], "gt_se_raw": [25.0, 30.0], "gt_se_norm": [-0.7773400545120239, -0.7328080534934998], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006918.pt"}
{"idx": 6920, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/13S9n41lme8.mp4", "duration": 224.55766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the camera focuses on the rough texture of a shark's skin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the camera focuses on the rough texture of a shark's skin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 118.0], "base_se_norm": [-0.06482815742492676, 0.05095505714416504], "gt_se_raw": [104.0, 118.0], "gt_se_norm": [-0.0737345814704895, 0.05095505714416504], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006920.pt"}
{"idx": 6922, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LjufgOcZLZs.mp4", "duration": 273.24, "prompt_text": "Please find the visual event described by the sentence 'When does the man explain the social experiment to the audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man explain the social experiment to the audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 49.0], "base_se_norm": [-0.8170107007026672, -0.641340970993042], "gt_se_raw": [25.0, 40.0], "gt_se_norm": [-0.8170107007026672, -0.7072170972824097], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006922.pt"}
{"idx": 6924, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LjufgOcZLZs.mp4", "duration": 273.24, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman wearing white pants points out the man gave her the wrong bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman wearing white pants points out the man gave her the wrong bill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 81.0], "base_se_norm": [-0.45835161209106445, -0.4071146249771118], "gt_se_raw": [74.0, 78.0], "gt_se_norm": [-0.45835161209106445, -0.4290733337402344], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006924.pt"}
{"idx": 6926, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LjufgOcZLZs.mp4", "duration": 273.24, "prompt_text": "Please find the visual event described by the sentence 'At what point does the experimenter reveal he is not blind to the man in the black tank top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 240.0 - 258.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the experimenter reveal he is not blind to the man in the black tank top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 240.0 - 258.0 seconds.<|im_end|>\n", "base_se_raw": [240.0, 258.0], "base_se_norm": [0.756697416305542, 0.8884497880935669], "gt_se_raw": [236.0, 243.0], "gt_se_norm": [0.7274191379547119, 0.7786562442779541], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006926.pt"}
{"idx": 6928, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4QdmRufojsU.mp4", "duration": 222.9562896229563, "prompt_text": "Please find the visual event described by the sentence 'Show the time when three people are doing lunges with punches.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when three people are doing lunges with punches.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [-0.3092816472053528, -0.25545942783355713], "gt_se_raw": [77.0, 83.0], "gt_se_norm": [-0.3092816472053528, -0.25545942783355713], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006928.pt"}
{"idx": 6930, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4QdmRufojsU.mp4", "duration": 222.9562896229563, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the segment where people are doing squats against the wall with an exercise ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the segment where people are doing squats against the wall with an exercise ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [-0.6142741441726685, -0.587363064289093], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [-0.6142741441726685, -0.587363064289093], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006930.pt"}
{"idx": 6932, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/98j7_2R5jdU.mp4", "duration": 320.0197, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man is preparing his equipment and says, 'Let's climb.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man is preparing his equipment and says, 'Let's climb.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 83.0], "base_se_norm": [-0.5000308156013489, -0.48128193616867065], "gt_se_raw": [78.0, 83.0], "gt_se_norm": [-0.5125300288200378, -0.48128193616867065], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006932.pt"}
{"idx": 6934, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/98j7_2R5jdU.mp4", "duration": 320.0197, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man working on a laptop at night, sitting on a ladder next to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 189.0 - 194.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man working on a laptop at night, sitting on a ladder next to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 189.0 - 194.0 seconds.<|im_end|>\n", "base_se_raw": [189.0, 194.0], "base_se_norm": [0.1811772584915161, 0.2124253511428833], "gt_se_raw": [190.0, 193.0], "gt_se_norm": [0.18742680549621582, 0.20617568492889404], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006934.pt"}
{"idx": 6936, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/98j7_2R5jdU.mp4", "duration": 320.0197, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a person is speaking in a classroom, writing on a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 200.0 - 213.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a person is speaking in a classroom, writing on a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 200.0 - 213.0 seconds.<|im_end|>\n", "base_se_raw": [200.0, 213.0], "base_se_norm": [0.2499229907989502, 0.33116793632507324], "gt_se_raw": [201.0, 205.0], "gt_se_norm": [0.25617265701293945, 0.2811710834503174], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006936.pt"}
{"idx": 6938, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n7N6_lIORNM.mp4", "duration": 270.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man is looking down at the memorial plaque.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man is looking down at the memorial plaque.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.6376186609268188, -0.608036458492279], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.6376186609268188, -0.6154320240020752], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006938.pt"}
{"idx": 6940, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n7N6_lIORNM.mp4", "duration": 270.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the three characters are standing together in front of the statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 228.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the three characters are standing together in front of the statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 228.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 228.0], "base_se_norm": [0.33859241008758545, 0.6861827373504639], "gt_se_raw": [181.0, 187.0], "gt_se_norm": [0.33859241008758545, 0.38296568393707275], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006940.pt"}
{"idx": 6942, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8j3NmTv9AWg.mp4", "duration": 252.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the punishment wheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the punishment wheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.7541714906692505, -0.7303816080093384], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.7541714906692505, -0.7303816080093384], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006942.pt"}
{"idx": 6944, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8j3NmTv9AWg.mp4", "duration": 252.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the grapefruit's head turn completely blue from the brain freeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the grapefruit's head turn completely blue from the brain freeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 100.0], "base_se_norm": [-0.25458449125289917, -0.20700478553771973], "gt_se_raw": [146.0, 148.0], "gt_se_norm": [0.15777301788330078, 0.17363297939300537], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006944.pt"}
{"idx": 6946, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8j3NmTv9AWg.mp4", "duration": 252.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is it announced that the grapefruit has lost the challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 206.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is it announced that the grapefruit has lost the challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 206.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 206.0], "base_se_norm": [0.6177102327346802, 0.6335701942443848], "gt_se_raw": [205.0, 207.0], "gt_se_norm": [0.6256401538848877, 0.6415001153945923], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006946.pt"}
{"idx": 6948, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAyVk_sH42k.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man show the boot-shaped glass to the camera for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man show the boot-shaped glass to the camera for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 74.0], "base_se_norm": [-0.32293397188186646, -0.2631928324699402], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [-0.312977135181427, -0.28310656547546387], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006948.pt"}
{"idx": 6950, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAyVk_sH42k.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man drinks from the boot-shaped glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man drinks from the boot-shaped glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 151.0], "base_se_norm": [0.3342183828353882, 0.503484845161438], "gt_se_raw": [134.0, 151.0], "gt_se_norm": [0.3342183828353882, 0.503484845161438], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006950.pt"}
{"idx": 6952, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SbQc_JLUH7k.mp4", "duration": 200.375, "prompt_text": "Please find the visual event described by the sentence 'When does the man use a phone app to shave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man use a phone app to shave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 68.0], "base_se_norm": [-0.5009357333183289, -0.321272611618042], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.4111042022705078, -0.35121649503707886], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006952.pt"}
{"idx": 6954, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SbQc_JLUH7k.mp4", "duration": 200.375, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man is looking for something in the drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man is looking for something in the drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 41.0], "base_se_norm": [-0.660636305809021, -0.5907673239707947], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [-0.6506550312042236, -0.5907673239707947], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006954.pt"}
{"idx": 6956, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SbQc_JLUH7k.mp4", "duration": 200.375, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where pictures of famous people are shown in a vintage mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where pictures of famous people are shown in a vintage mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 130.0], "base_se_norm": [0.12788522243499756, 0.2975670099258423], "gt_se_raw": [113.0, 131.0], "gt_se_norm": [0.12788522243499756, 0.3075484037399292], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006956.pt"}
{"idx": 6958, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8VkD__hXeM8.mp4", "duration": 323.64, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a fat man climbs into the top bunk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a fat man climbs into the top bunk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 68.0], "base_se_norm": [-0.6415770649909973, -0.5797799825668335], "gt_se_raw": [61.0, 68.0], "gt_se_norm": [-0.6230379343032837, -0.5797799825668335], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006958.pt"}
{"idx": 6960, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8VkD__hXeM8.mp4", "duration": 323.64, "prompt_text": "Please find the visual event described by the sentence 'A man sets a cloth on fire with his cigarette. Specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 150.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man sets a cloth on fire with his cigarette. Specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 150.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [150.0, 162.0], "base_se_norm": [-0.07304418087005615, 0.0011123418807983398], "gt_se_raw": [157.0, 162.0], "gt_se_norm": [-0.029786229133605957, 0.0011123418807983398], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006960.pt"}
{"idx": 6962, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S5z5_ukDxII.mp4", "duration": 204.6044, "prompt_text": "Please find the visual event described by the sentence 'Show me when the colorful paper stars are arranged in a circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the colorful paper stars are arranged in a circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.9218003153800964, -0.8533755540847778], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.9120253324508667, -0.8533755540847778], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006962.pt"}
{"idx": 6964, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S5z5_ukDxII.mp4", "duration": 204.6044, "prompt_text": "Please find the visual event described by the sentence 'When are the hands wrapping the long green paper strip around the pentagonal shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the hands wrapping the long green paper strip around the pentagonal shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 139.0], "base_se_norm": [-0.5894516706466675, 0.3587195873260498], "gt_se_raw": [66.0, 139.0], "gt_se_norm": [-0.354852557182312, 0.3587195873260498], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006964.pt"}
{"idx": 6966, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S5z5_ukDxII.mp4", "duration": 204.6044, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the five points of the star are being refined by pinching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the five points of the star are being refined by pinching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 168.0], "base_se_norm": [0.41736936569213867, 0.6421934366226196], "gt_se_raw": [165.0, 188.0], "gt_se_norm": [0.6128685474395752, 0.8376926183700562], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006966.pt"}
{"idx": 6968, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6azdnP1Dz0U.mp4", "duration": 212.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the mechanics and internal structure of the elevator are explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the mechanics and internal structure of the elevator are explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 50.0], "base_se_norm": [-0.6807011961936951, -0.5304429531097412], "gt_se_raw": [28.0, 52.0], "gt_se_norm": [-0.7370480298995972, -0.5116606950759888], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006968.pt"}
{"idx": 6970, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6azdnP1Dz0U.mp4", "duration": 212.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the elevator door is accidentally broken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the elevator door is accidentally broken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 112.0], "base_se_norm": [-0.013930141925811768, 0.05180788040161133], "gt_se_raw": [123.0, 132.0], "gt_se_norm": [0.15511035919189453, 0.23963069915771484], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006970.pt"}
{"idx": 6972, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/e9yPYbjA210.mp4", "duration": 277.277, "prompt_text": "Please find the visual event described by the sentence 'When does the man enter the facility and approach the front desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man enter the facility and approach the front desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 61.0], "base_se_norm": [-0.6898408532142639, -0.5600067973136902], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.7042668461799622, -0.6682018041610718], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006972.pt"}
{"idx": 6974, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/e9yPYbjA210.mp4", "duration": 277.277, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man performing a pulmonary function test by breathing into a device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man performing a pulmonary function test by breathing into a device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 122.0], "base_se_norm": [-0.19935661554336548, -0.12001359462738037], "gt_se_raw": [109.0, 121.0], "gt_se_norm": [-0.21378260850906372, -0.1272265911102295], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006974.pt"}
{"idx": 6976, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/e9yPYbjA210.mp4", "duration": 277.277, "prompt_text": "Please find the visual event described by the sentence 'When is the man submerged underwater for the hydrostatic weighing test?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 231.0 - 258.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man submerged underwater for the hydrostatic weighing test?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 231.0 - 258.0 seconds.<|im_end|>\n", "base_se_raw": [231.0, 258.0], "base_se_norm": [0.6662037372589111, 0.8609548807144165], "gt_se_raw": [251.0, 255.0], "gt_se_norm": [0.8104639053344727, 0.8393158912658691], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006976.pt"}
{"idx": 6978, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Slq2zeofV0w.mp4", "duration": 202.3021, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the movie give the woman flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the movie give the woman flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.31785184144973755, -0.2783070206642151], "gt_se_raw": [70.0, 74.0], "gt_se_norm": [-0.30796563625335693, -0.2684208154678345], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006978.pt"}
{"idx": 6980, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Slq2zeofV0w.mp4", "duration": 202.3021, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the woman in the film starts to cry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the woman in the film starts to cry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 146.0], "base_se_norm": [0.28520667552948, 0.4433859586715698], "gt_se_raw": [130.0, 140.0], "gt_se_norm": [0.28520667552948, 0.38406872749328613], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006980.pt"}
{"idx": 6982, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/80zhPlAesEE.mp4", "duration": 312.28, "prompt_text": "Please find the visual event described by the sentence 'When can we see the men on horseback riding through the water towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the men on horseback riding through the water towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.948763906955719], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9935954809188843, -0.9615729451179504], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006982.pt"}
{"idx": 6984, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/80zhPlAesEE.mp4", "duration": 312.28, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the foreman confronts Fantine and accuses her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 217.0 - 248.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the foreman confronts Fantine and accuses her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 217.0 - 248.0 seconds.<|im_end|>\n", "base_se_raw": [217.0, 248.0], "base_se_norm": [0.38977837562561035, 0.5883182287216187], "gt_se_raw": [262.0, 285.0], "gt_se_norm": [0.6779812574386597, 0.8252849578857422], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006984.pt"}
{"idx": 6986, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/80zhPlAesEE.mp4", "duration": 312.28, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the green coat enter the factory floor to break up the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 181.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the green coat enter the factory floor to break up the fight?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 181.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 181.0], "base_se_norm": [0.08236193656921387, 0.159216046333313], "gt_se_raw": [170.0, 177.0], "gt_se_norm": [0.08876645565032959, 0.13359808921813965], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006986.pt"}
{"idx": 6988, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DS8yeXFeEPA.mp4", "duration": 323.65666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where they are all jogging in place.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where they are all jogging in place.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.8207978010177612, -0.7651832699775696], "gt_se_raw": [16.0, 30.0], "gt_se_norm": [-0.9011297821998596, -0.8146183490753174], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006988.pt"}
{"idx": 6990, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DS8yeXFeEPA.mp4", "duration": 323.65666666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when they are all doing jumping jacks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when they are all doing jumping jacks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 144.0], "base_se_norm": [-0.14106512069702148, -0.1101682186126709], "gt_se_raw": [129.0, 140.0], "gt_se_norm": [-0.2028590440750122, -0.1348857879638672], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006990.pt"}
{"idx": 6992, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wWVppdfYOx8.mp4", "duration": 271.2, "prompt_text": "Please find the visual event described by the sentence 'Show me when the four corners of the paper are folded towards the center for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the four corners of the paper are folded towards the center for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 81.0], "base_se_norm": [-0.6976401209831238, -0.40265488624572754], "gt_se_raw": [41.0, 81.0], "gt_se_norm": [-0.6976401209831238, -0.40265488624572754], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006992.pt"}
{"idx": 6994, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wWVppdfYOx8.mp4", "duration": 271.2, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the paper is folded in half diagonally.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the paper is folded in half diagonally.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 23.0], "base_se_norm": [-0.9262536764144897, -0.8303834795951843], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.9115044474601746, -0.8893805146217346], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006994.pt"}
{"idx": 6996, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wWVppdfYOx8.mp4", "duration": 271.2, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for tearing the paper to create the dog's tail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 202.0 - 238.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for tearing the paper to create the dog's tail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 202.0 - 238.0 seconds.<|im_end|>\n", "base_se_raw": [202.0, 238.0], "base_se_norm": [0.4896754026412964, 0.7551621198654175], "gt_se_raw": [212.0, 217.0], "gt_se_norm": [0.5634217262268066, 0.6002949476242065], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006996.pt"}
{"idx": 6998, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SmSVlY2gx88.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where different types of disabilities are explained using pages from the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where different types of disabilities are explained using pages from the book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 55.0], "base_se_norm": [-0.6587681770324707, -0.5061118602752686], "gt_se_raw": [39.0, 56.0], "gt_se_norm": [-0.6497883796691895, -0.4971320629119873], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006998.pt"}
{"idx": 7000, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SmSVlY2gx88.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker read a passage about all children playing together in a pool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker read a passage about all children playing together in a pool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 188.0], "base_se_norm": [0.4547250270843506, 0.6881994009017944], "gt_se_raw": [161.0, 189.0], "gt_se_norm": [0.44574522972106934, 0.6971791982650757], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007000.pt"}
{"idx": 7002, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JfvEPc07UVY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'When are two men playing table tennis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men playing table tennis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [-0.6026073694229126, -0.5694912672042847], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.6026073694229126, -0.5694912672042847], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007002.pt"}
{"idx": 7004, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JfvEPc07UVY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where an elderly woman is flipping through a yearbook.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where an elderly woman is flipping through a yearbook.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 81.0], "base_se_norm": [-0.39563196897506714, -0.32939988374710083], "gt_se_raw": [73.0, 81.0], "gt_se_norm": [-0.39563196897506714, -0.32939988374710083], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007004.pt"}
{"idx": 7006, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JfvEPc07UVY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where mannequins wearing various dresses are displayed in a museum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 183.0 - 203.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where mannequins wearing various dresses are displayed in a museum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 183.0 - 203.0 seconds.<|im_end|>\n", "base_se_raw": [183.0, 203.0], "base_se_norm": [0.5150595903396606, 0.6806398630142212], "gt_se_raw": [183.0, 186.0], "gt_se_norm": [0.5150595903396606, 0.5398966073989868], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007006.pt"}
{"idx": 7008, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1Ez6dw3ywcc.mp4", "duration": 207.70749999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man in the blue suit first enters the room where the woman is.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man in the blue suit first enters the room where the woman is.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 62.0], "base_se_norm": [-0.46078020334243774, -0.4030066728591919], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [-0.43189340829849243, -0.41263556480407715], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007008.pt"}
{"idx": 7010, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1Ez6dw3ywcc.mp4", "duration": 207.70749999999998, "prompt_text": "Please find the visual event described by the sentence 'When are the police officers seen talking to the couple inside the apartment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 165.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the police officers seen talking to the couple inside the apartment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 165.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 165.0], "base_se_norm": [0.36730730533599854, 0.5887726545333862], "gt_se_raw": [141.0, 164.0], "gt_se_norm": [0.3576784133911133, 0.579143762588501], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007010.pt"}
{"idx": 7012, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vqDbMEdLiCs.mp4", "duration": 238.96, "prompt_text": "Please find the visual event described by the sentence 'When does the man hold a silver hard drive and a book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man hold a silver hard drive and a book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.9414128065109253, -0.8911951780319214], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.9246735572814941, -0.899564802646637], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007012.pt"}
{"idx": 7014, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vqDbMEdLiCs.mp4", "duration": 238.96, "prompt_text": "Please find the visual event described by the sentence 'At what point is the temperature of the book measured with an infrared thermometer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the temperature of the book measured with an infrared thermometer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 72.0], "base_se_norm": [-0.4643455147743225, -0.3973886966705322], "gt_se_raw": [64.0, 69.0], "gt_se_norm": [-0.4643455147743225, -0.4224975109100342], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007014.pt"}
{"idx": 7016, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vqDbMEdLiCs.mp4", "duration": 238.96, "prompt_text": "Please find the visual event described by the sentence 'When does the host explain the concept of thermal conductivity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the host explain the concept of thermal conductivity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 126.0], "base_se_norm": [-0.05423504114151001, 0.05456972122192383], "gt_se_raw": [162.0, 166.0], "gt_se_norm": [0.3558753728866577, 0.38935387134552], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007016.pt"}
{"idx": 7018, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ikGVWEvUzNM.mp4", "duration": 318.2083333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the sequence where the woman falls to the ground in the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the sequence where the woman falls to the ground in the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.6794552803039551, -0.6480293273925781], "gt_se_raw": [51.0, 55.0], "gt_se_norm": [-0.6794552803039551, -0.6543145179748535], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007018.pt"}
{"idx": 7020, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ikGVWEvUzNM.mp4", "duration": 318.2083333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the girl takes off the mask and reveals her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 257.0 - 262.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the girl takes off the mask and reveals her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 257.0 - 262.0 seconds.<|im_end|>\n", "base_se_raw": [257.0, 262.0], "base_se_norm": [0.6152938604354858, 0.6467198133468628], "gt_se_raw": [256.0, 261.0], "gt_se_norm": [0.6090086698532104, 0.6404346227645874], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007020.pt"}
{"idx": 7022, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTb-DIU9vj8.mp4", "duration": 267.6840833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher show a picture of an earwig on the projector?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher show a picture of an earwig on the projector?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 22.0], "base_se_norm": [-0.8953990936279297, -0.8356271386146545], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.8356271386146545, -0.8206841349601746], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007022.pt"}
{"idx": 7024, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTb-DIU9vj8.mp4", "duration": 267.6840833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the bully puts his wet finger into the other boy's ear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the bully puts his wet finger into the other boy's ear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 59.0], "base_se_norm": [-0.589067816734314, -0.5591818690299988], "gt_se_raw": [55.0, 59.0], "gt_se_norm": [-0.589067816734314, -0.5591818690299988], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007024.pt"}
{"idx": 7026, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTb-DIU9vj8.mp4", "duration": 267.6840833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy with red hair's eyes turn completely white.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 183.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy with red hair's eyes turn completely white.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 183.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [183.0, 188.0], "base_se_norm": [0.36728334426879883, 0.40464091300964355], "gt_se_raw": [184.0, 188.0], "gt_se_norm": [0.3747549057006836, 0.40464091300964355], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007026.pt"}
{"idx": 7028, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JuKCOthud68.mp4", "duration": 217.3838333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a cassette is being loaded into a duplicator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a cassette is being loaded into a duplicator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 94.0], "base_se_norm": [-0.15357089042663574, -0.13517028093338013], "gt_se_raw": [92.0, 94.0], "gt_se_norm": [-0.15357089042663574, -0.13517028093338013], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007028.pt"}
{"idx": 7030, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JuKCOthud68.mp4", "duration": 217.3838333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman helps a man with a paper-mache project.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman helps a man with a paper-mache project.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 149.0], "base_se_norm": [0.35244643688201904, 0.37084710597991943], "gt_se_raw": [147.0, 149.0], "gt_se_norm": [0.35244643688201904, 0.37084710597991943], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007030.pt"}
{"idx": 7032, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vHjv17F8iiE.mp4", "duration": 201.9, "prompt_text": "Please find the visual event described by the sentence 'At what point is the player in the pink checkered shirt seen killing another player in a cave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the player in the pink checkered shirt seen killing another player in a cave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.7226349711418152, -0.6830114126205444], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.7127290964126587, -0.6830114126205444], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007032.pt"}
{"idx": 7034, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vHjv17F8iiE.mp4", "duration": 201.9, "prompt_text": "Please find the visual event described by the sentence 'When does the player with a pig head in a suit get killed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player with a pig head in a suit get killed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 156.0], "base_se_norm": [0.4957900047302246, 0.5453195571899414], "gt_se_raw": [155.0, 157.0], "gt_se_norm": [0.5354136228561401, 0.5552253723144531], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007034.pt"}
{"idx": 7036, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vHjv17F8iiE.mp4", "duration": 201.9, "prompt_text": "Please find the visual event described by the sentence 'When does the player in the purple shirt start placing note blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in the purple shirt start placing note blocks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 13.0], "base_se_norm": [-0.9405646324157715, -0.8712233901023865], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.930658757686615, -0.9108469486236572], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007036.pt"}
{"idx": 7038, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFWmYhkMB5Y.mp4", "duration": 327.16016666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two men are looking at a photo album and see a white faceless figure in a suit in the photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two men are looking at a photo album and see a white faceless figure in a suit in the photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.7738111019134521, -0.7554714679718018], "gt_se_raw": [24.0, 31.0], "gt_se_norm": [-0.8532828688621521, -0.8104903697967529], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007038.pt"}
{"idx": 7040, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFWmYhkMB5Y.mp4", "duration": 327.16016666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the white mask reveal that he is not Slender Man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 250.0 - 262.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the white mask reveal that he is not Slender Man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 250.0 - 262.0 seconds.<|im_end|>\n", "base_se_raw": [250.0, 262.0], "base_se_norm": [0.5283035039901733, 0.6016620397567749], "gt_se_raw": [216.0, 220.0], "gt_se_norm": [0.3204542398452759, 0.3449070453643799], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007040.pt"}
{"idx": 7042, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g0FMrjsYAzc.mp4", "duration": 318.61830000000003, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a person is sitting in a helicopter looking out the window at the snowy mountains.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a person is sitting in a helicopter looking out the window at the snowy mountains.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 27.0], "base_se_norm": [-0.8932892084121704, -0.8305181860923767], "gt_se_raw": [17.0, 27.0], "gt_se_norm": [-0.8932892084121704, -0.8305181860923767], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007042.pt"}
{"idx": 7044, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g0FMrjsYAzc.mp4", "duration": 318.61830000000003, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see a skier skiing on a wide, snowy plain under the bright sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see a skier skiing on a wide, snowy plain under the bright sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 85.0], "base_se_norm": [-0.5166630148887634, -0.4664462208747864], "gt_se_raw": [77.0, 85.0], "gt_se_norm": [-0.5166630148887634, -0.4664462208747864], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007044.pt"}
{"idx": 7046, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g0FMrjsYAzc.mp4", "duration": 318.61830000000003, "prompt_text": "Please find the visual event described by the sentence 'When can we see the skier making their way down a steep, narrow snow ridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the skier making their way down a steep, narrow snow ridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 68.0], "base_se_norm": [-0.6673135161399841, -0.5731569528579712], "gt_se_raw": [226.0, 245.0], "gt_se_norm": [0.41862547397613525, 0.5378904342651367], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007046.pt"}
{"idx": 1, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ro7tuDVP6Ks.mp4", "duration": 82.0825, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man in a suit is speaking at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man in a suit is speaking at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [-0.48831963539123535, -0.43958818912506104], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.5126854181289673, -0.43958818912506104], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000001.pt"}
{"idx": 3, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ro7tuDVP6Ks.mp4", "duration": 82.0825, "prompt_text": "Please find the visual event described by the sentence 'Show me when the students are cheering in the auditorium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the students are cheering in the auditorium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 66.0], "base_se_norm": [0.43757808208465576, 0.6081382036209106], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [0.4619438648223877, 0.510675311088562], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000003.pt"}
{"idx": 5, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdwFR_74t_0.mp4", "duration": 12.654, "prompt_text": "Please find the visual event described by the sentence 'When does the boy pull the blue container from under the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy pull the blue container from under the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.6838943958282471, -0.20973604917526245], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.6838943958282471, -0.20973604917526245], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000005.pt"}
{"idx": 7, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdwFR_74t_0.mp4", "duration": 12.654, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the boy picks up and holds the large, flat game board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the boy picks up and holds the large, flat game board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.20973604917526245, 0.8966333866119385], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [0.1063694953918457, 0.7385807037353516], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000007.pt"}
{"idx": 9, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdwFR_74t_0.mp4", "duration": 12.654, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the boy reaches into the blue box while it is still under the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the boy reaches into the blue box while it is still under the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.3677888512611389], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.6838943958282471], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000009.pt"}
{"idx": 11, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BeiRM6-PdOM.mp4", "duration": 335.923078, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man and his instructor jump out of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 152.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man and his instructor jump out of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 152.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 152.0], "base_se_norm": [-0.19029080867767334, -0.09503090381622314], "gt_se_raw": [141.0, 144.0], "gt_se_norm": [-0.16052204370498657, -0.14266085624694824], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000011.pt"}
{"idx": 13, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BeiRM6-PdOM.mp4", "duration": 335.923078, "prompt_text": "Please find the visual event described by the sentence 'At what point is the parachute deployed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 242.0 - 245.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the parachute deployed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 242.0 - 245.0 seconds.<|im_end|>\n", "base_se_raw": [242.0, 245.0], "base_se_norm": [0.44080615043640137, 0.4586672782897949], "gt_se_raw": [242.0, 245.0], "gt_se_norm": [0.44080615043640137, 0.4586672782897949], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000013.pt"}
{"idx": 15, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jIi4Fe6jRHo.mp4", "duration": 57.771244, "prompt_text": "Please find the visual event described by the sentence 'A man runs towards a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man runs towards a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8615227937698364], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.8269035220146179], "start_token_span": [61, 64], "end_token_span": [66, 69], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000015.pt"}
{"idx": 17, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jIi4Fe6jRHo.mp4", "duration": 57.771244, "prompt_text": "Please find the visual event described by the sentence 'A shirtless man is drawing a bowstring in front of a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A shirtless man is drawing a bowstring in front of a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 39.0], "base_se_norm": [-0.030659615993499756, 0.35015273094177246], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [0.00395965576171875, 0.07319831848144531], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000017.pt"}
{"idx": 19, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jIi4Fe6jRHo.mp4", "duration": 57.771244, "prompt_text": "Please find the visual event described by the sentence 'The man shoots an explosive arrow, causing a big explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The man shoots an explosive arrow, causing a big explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [0.6963456869125366, 0.8348228931427002], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.6617263555526733, 0.8002035617828369], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000019.pt"}
{"idx": 21, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pK-7v-CimFo.mp4", "duration": 60.07, "prompt_text": "Please find the visual event described by the sentence 'When is the person squeezing the water out of the shredded potatoes using a cheesecloth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person squeezing the water out of the shredded potatoes using a cheesecloth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 19.0], "base_se_norm": [-0.5671716332435608, -0.3674046993255615], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.5671716332435608, -0.46728813648223877], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000021.pt"}
{"idx": 23, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pK-7v-CimFo.mp4", "duration": 60.07, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the finished waffle is being garnished with scallions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the finished waffle is being garnished with scallions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 52.0], "base_se_norm": [0.6647244691848755, 0.7313134670257568], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [0.6647244691848755, 0.7313134670257568], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000023.pt"}
{"idx": 25, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x9FzWnWW95U.mp4", "duration": 10.541797, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman with purple hair and butterfly-like glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 1.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman with purple hair and butterfly-like glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 1.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 1.0], "base_se_norm": [-1.0, -0.8102790117263794], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8102790117263794], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000025.pt"}
{"idx": 27, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x9FzWnWW95U.mp4", "duration": 10.541797, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a blonde woman in a purple top is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a blonde woman in a purple top is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [0.5177677869796753, 0.8972097635269165], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [0.5177677869796753, 0.8972097635269165], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000027.pt"}
{"idx": 29, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x9FzWnWW95U.mp4", "duration": 10.541797, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a woman wearing a large gold necklace is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 1.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a woman wearing a large gold necklace is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 1.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 1.0], "base_se_norm": [-1.0, -0.8102790117263794], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8102790117263794], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000029.pt"}
{"idx": 31, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yH2wT5aTmb8.mp4", "duration": 31.833333, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the woman is holding and moving the baby's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the woman is holding and moving the baby's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 9.0], "base_se_norm": [-0.8743455410003662, -0.43455493450164795], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.8115183115005493, -0.49738216400146484], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000031.pt"}
{"idx": 33, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yH2wT5aTmb8.mp4", "duration": 31.833333, "prompt_text": "Please find the visual event described by the sentence 'Identify the period where the baby is sitting still, looking towards the blanket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the period where the baby is sitting still, looking towards the blanket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 23.0], "base_se_norm": [-0.05759155750274658, 0.4450262784957886], "gt_se_raw": [17.0, 24.0], "gt_se_norm": [0.06806290149688721, 0.5078535079956055], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000033.pt"}
{"idx": 35, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qniwI2hNhDs.mp4", "duration": 165.163333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when two dogs are running side-by-side on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when two dogs are running side-by-side on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.6367232799530029, -0.612504780292511], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.6367232799530029, -0.612504780292511], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000035.pt"}
{"idx": 37, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qniwI2hNhDs.mp4", "duration": 165.163333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where several men are grabbing sheep inside a pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where several men are grabbing sheep inside a pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.4066479802131653, -0.35821104049682617], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.38242948055267334, -0.35821104049682617], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000037.pt"}
{"idx": 39, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qniwI2hNhDs.mp4", "duration": 165.163333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the sheep with lights form the Mona Lisa painting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the sheep with lights form the Mona Lisa painting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 122.0], "base_se_norm": [0.41677939891815186, 0.47732555866241455], "gt_se_raw": [118.0, 122.0], "gt_se_norm": [0.42888855934143066, 0.47732555866241455], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000039.pt"}
{"idx": 41, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uwFhl9lG-z0.mp4", "duration": 443.826211, "prompt_text": "Please find the visual event described by the sentence 'At what point does she discuss the limitations caused by not being able to drive?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she discuss the limitations caused by not being able to drive?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [-0.5493731498718262, -0.495297908782959], "gt_se_raw": [52.0, 69.0], "gt_se_norm": [-0.7656740546226501, -0.6890674829483032], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000041.pt"}
{"idx": 43, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uwFhl9lG-z0.mp4", "duration": 443.826211, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker give an example of someone speaking to her companion instead of her directly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker give an example of someone speaking to her companion instead of her directly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 122.0], "base_se_norm": [-0.49980419874191284, -0.4502352476119995], "gt_se_raw": [351.0, 372.0], "gt_se_norm": [0.5817002058029175, 0.6763318777084351], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000043.pt"}
{"idx": 45, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-CR65sS1Frw.mp4", "duration": 5.410233, "prompt_text": "Please find the visual event described by the sentence 'When does the initial bright flash of the explosion happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 1.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the initial bright flash of the explosion happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 1.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 1.0], "base_se_norm": [-1.0, -0.6303301453590393], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.6303301453590393], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000045.pt"}
{"idx": 47, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-CR65sS1Frw.mp4", "duration": 5.410233, "prompt_text": "Please find the visual event described by the sentence 'At what point does the fireball turn into a large plume of black smoke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the fireball turn into a large plume of black smoke?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 3.0], "base_se_norm": [-0.2606602907180786, 0.1090095043182373], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.2606602907180786, 0.1090095043182373], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000047.pt"}
{"idx": 49, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-CR65sS1Frw.mp4", "duration": 5.410233, "prompt_text": "Please find the visual event described by the sentence 'Identify the entire duration of the explosion event, from the initial spark to the end of the clip.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the entire duration of the explosion event, from the initial spark to the end of the clip.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, 0.1090095043182373], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, 0.4786794185638428], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000049.pt"}
{"idx": 51, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KxwbAqLb5Qk.mp4", "duration": 105.743, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display a list of the shell's characteristics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display a list of the shell's characteristics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.7162932753562927, -0.6595519185066223], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.7162932753562927, -0.6595519185066223], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000051.pt"}
{"idx": 53, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KxwbAqLb5Qk.mp4", "duration": 105.743, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of the tortoise hiding under a rock crevice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of the tortoise hiding under a rock crevice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.07322466373443604, -0.035397112369537354], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.07322466373443604, -0.035397112369537354], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000053.pt"}
{"idx": 55, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iimJIWQqAGA.mp4", "duration": 58.04, "prompt_text": "Please find the visual event described by the sentence 'When does the girl put a coin into the vending machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl put a coin into the vending machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.5864920616149902, -0.41419708728790283], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.5864920616149902, -0.5175740718841553], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000055.pt"}
{"idx": 57, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iimJIWQqAGA.mp4", "duration": 58.04, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the pink balloon comes out of the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the pink balloon comes out of the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 30.0], "base_se_norm": [-0.2419021725654602, 0.033769845962524414], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.2763611674308777, -0.17298418283462524], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000057.pt"}
{"idx": 59, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iimJIWQqAGA.mp4", "duration": 58.04, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl walking away with the pink heart-shaped balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl walking away with the pink heart-shaped balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 44.0], "base_se_norm": [0.3094416856765747, 0.5161956548690796], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [0.3094416856765747, 0.4817366600036621], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000059.pt"}
{"idx": 61, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3xPDE8H0Jw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where two young men are installing materials on the roof's edge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where two young men are installing materials on the roof's edge.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000061.pt"}
{"idx": 63, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3xPDE8H0Jw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment an instructor is supervising a student marking a wooden plank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment an instructor is supervising a student marking a wooden plank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000063.pt"}
{"idx": 65, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4Pr7PZ3Bgc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see several students working together at a table with evidence markers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see several students working together at a table with evidence markers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.6060894131660461], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000065.pt"}
{"idx": 67, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4Pr7PZ3Bgc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man is lecturing in front of a class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man is lecturing in front of a class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [0.18173182010650635, 0.3786870241165161], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000067.pt"}
{"idx": 69, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4Pr7PZ3Bgc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman in a red shirt is teaching a group of students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman in a red shirt is teaching a group of students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.8686964511871338, -0.540437638759613], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.8686964511871338, -0.6717411875724792], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000069.pt"}
{"idx": 71, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RsZE4a4ol_U.mp4", "duration": 129.06226666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a dishwasher is opened.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a dishwasher is opened.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 101.0], "base_se_norm": [0.5186467170715332, 0.5651359558105469], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [0.5186467170715332, 0.5651359558105469], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000071.pt"}
{"idx": 73, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RsZE4a4ol_U.mp4", "duration": 129.06226666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the segment that introduces the '6-Person Suite'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment that introduces the '6-Person Suite'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 25.0], "base_se_norm": [-0.7210649251937866, -0.6125900745391846], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.7055684924125671, -0.628086507320404], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000073.pt"}
{"idx": 75, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VyHV0BRtdxo.mp4", "duration": 136.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a boy successfully summons his broom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a boy successfully summons his broom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 58.0], "base_se_norm": [-0.26583051681518555, -0.1483634114265442], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.26583051681518555, -0.236463725566864], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000075.pt"}
{"idx": 77, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VyHV0BRtdxo.mp4", "duration": 136.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the scene where the three-headed dog is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the scene where the three-headed dog is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 114.0], "base_se_norm": [0.6445397138595581, 0.6739064455032349], "gt_se_raw": [112.0, 114.0], "gt_se_norm": [0.6445397138595581, 0.6739064455032349], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000077.pt"}
{"idx": 79, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VyHV0BRtdxo.mp4", "duration": 136.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the students are arriving at Hogwarts by boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the students are arriving at Hogwarts by boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.7063322067260742, -0.5888650417327881], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [-0.6916488409042358, -0.5888650417327881], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000079.pt"}
{"idx": 81, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9-k5J4RxQdE.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'When does Daffy's beak get blown to the back of his head for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Daffy's beak get blown to the back of his head for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.23076921701431274, -0.13846153020858765], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.26153844594955444, -0.19999998807907104], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000081.pt"}
{"idx": 83, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9-k5J4RxQdE.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Show the moment Daffy's beak is blown off by the second shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment Daffy's beak is blown off by the second shot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 49.0], "base_se_norm": [0.4461538791656494, 0.5076923370361328], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.4461538791656494, 0.5076923370361328], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000083.pt"}
{"idx": 85, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/skKUzMST92g.mp4", "duration": 144.0, "prompt_text": "Please find the visual event described by the sentence 'When does the video introduce the tutorial's topic: 2x2 ribbing on double pointed needles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video introduce the tutorial's topic: 2x2 ribbing on double pointed needles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 12.0], "base_se_norm": [-0.9583333134651184, -0.8333333134651184], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.9027777910232544, -0.8333333134651184], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000085.pt"}
{"idx": 87, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/skKUzMST92g.mp4", "duration": 144.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the knitter demonstrate how to switch to the next needle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the knitter demonstrate how to switch to the next needle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 59.0], "base_se_norm": [-0.4444444179534912, -0.1805555820465088], "gt_se_raw": [40.0, 47.0], "gt_se_norm": [-0.4444444179534912, -0.3472222089767456], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000087.pt"}
{"idx": 89, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/skKUzMST92g.mp4", "duration": 144.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the knitter first uses a wooden ruler to measure the work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the knitter first uses a wooden ruler to measure the work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 109.0], "base_se_norm": [0.2916666269302368, 0.5138888359069824], "gt_se_raw": [92.0, 100.0], "gt_se_norm": [0.2777777910232544, 0.3888888359069824], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000089.pt"}
{"idx": 91, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xn_XpOYgUi8.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of simmering the tea leaves in the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of simmering the tea leaves in the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [-0.10138481855392456, 0.11719727516174316], "gt_se_raw": [41.0, 45.0], "gt_se_norm": [-0.004237234592437744, 0.09291040897369385], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000091.pt"}
{"idx": 93, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xn_XpOYgUi8.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the final tea poured through a strainer into the cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the final tea poured through a strainer into the cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 70.0], "base_se_norm": [0.5786483287811279, 0.7000827789306641], "gt_se_raw": [65.0, 69.0], "gt_se_norm": [0.5786483287811279, 0.6757959127426147], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000093.pt"}
{"idx": 95, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vg3Q2W-oGgE.mp4", "duration": 147.01353333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is organizing the cards on the desk while standing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is organizing the cards on the desk while standing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 34.0], "base_se_norm": [-0.9455832242965698, -0.5374575853347778], "gt_se_raw": [4.0, 23.0], "gt_se_norm": [-0.9455832242965698, -0.6871036291122437], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000095.pt"}
{"idx": 97, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vg3Q2W-oGgE.mp4", "duration": 147.01353333333333, "prompt_text": "Please find the visual event described by the sentence 'When does she take her craft project out of the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does she take her craft project out of the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 97.0], "base_se_norm": [0.02031421661376953, 0.3196064233779907], "gt_se_raw": [73.0, 96.0], "gt_se_norm": [-0.006894171237945557, 0.30600225925445557], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000097.pt"}
{"idx": 99, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vg3Q2W-oGgE.mp4", "duration": 147.01353333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman sitting in the chair and wrapping yarn around the sticks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman sitting in the chair and wrapping yarn around the sticks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 146.0], "base_se_norm": [0.6052943468093872, 0.9862116575241089], "gt_se_raw": [125.0, 147.0], "gt_se_norm": [0.7005237340927124, 0.999815821647644], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000099.pt"}
{"idx": 101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-E3cRZCne4.mp4", "duration": 75.04, "prompt_text": "Please find the visual event described by the sentence 'At what point is the phrase 'A Gallup World Poll' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the phrase 'A Gallup World Poll' shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.22707891464233398, -0.09381663799285889], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.22707891464233398, -0.17377400398254395], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000101.pt"}
{"idx": 103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-E3cRZCne4.mp4", "duration": 75.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that says 'Everyone is entitled to decent work'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that says 'Everyone is entitled to decent work'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [0.3859274387359619, 0.465884804725647], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [0.3859274387359619, 0.465884804725647], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000103.pt"}
{"idx": 105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_ikKgIuqVX4.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9030095934867859, -0.32106727361679077], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9030095934867859, -0.4180576801300049], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000105.pt"}
{"idx": 107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_ikKgIuqVX4.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention the phrase 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention the phrase 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.030096113681793213, 0.16388463973999023], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.0668942928314209, 0.2608751058578491], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000107.pt"}
{"idx": 109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_ikKgIuqVX4.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the phrase 'people with disabilities employed' is spoken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the phrase 'people with disabilities employed' is spoken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.12708652019500732, 0.0668942928314209], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000109.pt"}
{"idx": 111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QnrNPVaeDMk.mp4", "duration": 100.13336666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man and boy interact with the toothbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man and boy interact with the toothbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 37.0], "base_se_norm": [-0.7802929878234863, -0.2609856128692627], "gt_se_raw": [26.0, 37.0], "gt_se_norm": [-0.48069262504577637, -0.2609856128692627], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000111.pt"}
{"idx": 113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QnrNPVaeDMk.mp4", "duration": 100.13336666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is brushing the boy's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is brushing the boy's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 20.0], "base_se_norm": [-0.8601864576339722, -0.6005327701568604], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [0.07856154441833496, 0.1584550142288208], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000113.pt"}
{"idx": 115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dev6T3ZCrY8.mp4", "duration": 34.034, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the aerial shot of a frozen lake between snowy mountains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the aerial shot of a frozen lake between snowy mountains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.7649409770965576, -0.6474114060401917], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8237056732177734, -0.7061761617660522], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000115.pt"}
{"idx": 117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dev6T3ZCrY8.mp4", "duration": 34.034, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man in a beanie hat is in a boat on a rushing river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man in a beanie hat is in a boat on a rushing river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.2948228120803833, -0.11852854490280151], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.2360580563545227, -0.11852854490280151], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000117.pt"}
{"idx": 119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dev6T3ZCrY8.mp4", "duration": 34.034, "prompt_text": "Please find the visual event described by the sentence 'When does the title 'THE CALL OF THE WILD' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the title 'THE CALL OF THE WILD' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [0.762942910194397, 0.8804724216461182], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [0.762942910194397, 0.8804724216461182], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000119.pt"}
{"idx": 121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EMJ0kz8Iu9o.mp4", "duration": 54.320933333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the period when the concrete frame of the building is being constructed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the period when the concrete frame of the building is being constructed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 43.0], "base_se_norm": [-0.0059081315994262695, 0.5831834077835083], "gt_se_raw": [19.0, 41.0], "gt_se_norm": [-0.3004538416862488, 0.5095469951629639], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000121.pt"}
{"idx": 123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EMJ0kz8Iu9o.mp4", "duration": 54.320933333333336, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the building's exterior facade is being installed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the building's exterior facade is being installed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 49.0], "base_se_norm": [0.5831834077835083, 0.8040926456451416], "gt_se_raw": [44.0, 49.0], "gt_se_norm": [0.6200015544891357, 0.8040926456451416], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000123.pt"}
{"idx": 125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3_TG_mMywxE.mp4", "duration": 89.54779166666665, "prompt_text": "Please find the visual event described by the sentence 'When are the two presenters introducing themselves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two presenters introducing themselves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.7766555547714233, -0.6873177886009216], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.7766555547714233, -0.6873177886009216], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000125.pt"}
{"idx": 127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3_TG_mMywxE.mp4", "duration": 89.54779166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is the real-time drill depth displayed on the wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the real-time drill depth displayed on the wood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 44.0], "base_se_norm": [-0.12895673513412476, -0.017284512519836426], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [-0.15129119157791138, -0.03961896896362305], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000127.pt"}
{"idx": 129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3_TG_mMywxE.mp4", "duration": 89.54779166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the saw is being used for a precision cut, as indicated by the text on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the saw is being used for a precision cut, as indicated by the text on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 72.0], "base_se_norm": [0.38473546504974365, 0.6080799102783203], "gt_se_raw": [63.0, 67.0], "gt_se_norm": [0.4070699214935303, 0.4964076280593872], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000129.pt"}
{"idx": 131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RzR_O2DoSVs.mp4", "duration": 29.229200000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the child is covering their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the child is covering their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5737687349319458, 0.9158923625946045], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.5053439140319824, 0.9158923625946045], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000131.pt"}
{"idx": 133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RzR_O2DoSVs.mp4", "duration": 29.229200000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the adult raise their hand, prompting the child to raise their own hand to their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the adult raise their hand, prompting the child to raise their own hand to their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 18.0], "base_se_norm": [-0.6578763723373413, 0.2316451072692871], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [0.026370882987976074, 0.4369192123413086], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000133.pt"}
{"idx": 135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/18oCRSIVeec.mp4", "duration": 101.20240480961924, "prompt_text": "Please find the visual event described by the sentence 'When does the student put on the smart glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the student put on the smart glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 25.0], "base_se_norm": [-0.6245148181915283, -0.5059406161308289], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.6245148181915283, -0.5849901437759399], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000135.pt"}
{"idx": 137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/18oCRSIVeec.mp4", "duration": 101.20240480961924, "prompt_text": "Please find the visual event described by the sentence 'At what point does the student use his phone to interact with the vending machine's payment system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the student use his phone to interact with the vending machine's payment system?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [0.00788116455078125, 0.08693063259124756], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [0.00788116455078125, 0.0671682357788086], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000137.pt"}
{"idx": 139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/18oCRSIVeec.mp4", "duration": 101.20240480961924, "prompt_text": "Please find the visual event described by the sentence 'When can we see the student retrieve the can from the vending machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the student retrieve the can from the vending machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 78.0], "base_se_norm": [0.40312862396240234, 0.5414652824401855], "gt_se_raw": [73.0, 76.0], "gt_se_norm": [0.4426534175872803, 0.5019406080245972], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000139.pt"}
{"idx": 141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_iU_gnn28U.mp4", "duration": 190.12326666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the protagonist is called 'the Ghost' by the enemy leader.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the protagonist is called 'the Ghost' by the enemy leader.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 93.0], "base_se_norm": [-0.21103817224502563, -0.021687328815460205], "gt_se_raw": [82.0, 87.0], "gt_se_norm": [-0.13740170001983643, -0.0848042368888855], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000141.pt"}
{"idx": 143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_iU_gnn28U.mp4", "duration": 190.12326666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the protagonist claims 'Honor died on the beach'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the protagonist claims 'Honor died on the beach'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 93.0], "base_se_norm": [-0.06376528739929199, -0.021687328815460205], "gt_se_raw": [128.0, 131.0], "gt_se_norm": [0.3464949131011963, 0.3780534267425537], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000143.pt"}
{"idx": 145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MizXqgEuhe4.mp4", "duration": 77.04363333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'Money Handling Tips for Visually Impaired Youth' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'Money Handling Tips for Visually Impaired Youth' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.974040687084198, -0.922122061252594], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.922122061252594], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000145.pt"}
{"idx": 147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MizXqgEuhe4.mp4", "duration": 77.04363333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.4029357433319092, -0.27313917875289917], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.4288950562477112, -0.27313917875289917], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000147.pt"}
{"idx": 149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MizXqgEuhe4.mp4", "duration": 77.04363333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the Braille Institute logo shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the Braille Institute logo shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 76.0], "base_se_norm": [0.8950300216674805, 0.9729080200195312], "gt_se_raw": [73.0, 76.0], "gt_se_norm": [0.8950300216674805, 0.9729080200195312], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000149.pt"}
{"idx": 151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKIIf6W16Uk.mp4", "duration": 276.6433099766433, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man in black stands among giant gummy bear statues.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man in black stands among giant gummy bear statues.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 90.0], "base_se_norm": [-0.3782607913017273, -0.3493426442146301], "gt_se_raw": [86.0, 90.0], "gt_se_norm": [-0.3782607913017273, -0.3493426442146301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000151.pt"}
{"idx": 153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKIIf6W16Uk.mp4", "duration": 276.6433099766433, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is lying in a pool of colorful sprinkles and making a sprinkle angel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 211.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is lying in a pool of colorful sprinkles and making a sprinkle angel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 211.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [211.0, 222.0], "base_se_norm": [0.5254299640655518, 0.604954719543457], "gt_se_raw": [211.0, 222.0], "gt_se_norm": [0.5254299640655518, 0.604954719543457], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000153.pt"}
{"idx": 155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VfDWQG47pAQ.mp4", "duration": 177.24373333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a black suit holds a dog in a briefcase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a black suit holds a dog in a briefcase.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.9097288250923157, -0.8871610164642334], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [-0.9097288250923157, -0.8984449505805969], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000155.pt"}
{"idx": 157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VfDWQG47pAQ.mp4", "duration": 177.24373333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man in a black shirt stops the bullets with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man in a black shirt stops the bullets with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 118.0], "base_se_norm": [0.21866083145141602, 0.3314998149871826], "gt_se_raw": [108.0, 117.0], "gt_se_norm": [0.21866083145141602, 0.32021594047546387], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000157.pt"}
{"idx": 159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VfDWQG47pAQ.mp4", "duration": 177.24373333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where two people kiss at a bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where two people kiss at a bar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 167.0], "base_se_norm": [0.8618429899215698, 0.8844108581542969], "gt_se_raw": [165.0, 167.0], "gt_se_norm": [0.8618429899215698, 0.8844108581542969], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000159.pt"}
{"idx": 161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_WrjdMns5Vc.mp4", "duration": 19.227541666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the student's hand filling in the scantron sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the student's hand filling in the scantron sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.6879476308822632, 0.35222697257995605], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.3758952021598816, -0.16786032915115356], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000161.pt"}
{"idx": 163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_WrjdMns5Vc.mp4", "duration": 19.227541666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the bell ring to signal the end of the test?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bell ring to signal the end of the test?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.06384283304214478, 0.14419209957122803], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.35222697257995605, 0.5602619647979736], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000163.pt"}
{"idx": 165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-Z8xxygd2Y.mp4", "duration": 105.76, "prompt_text": "Please find the visual event described by the sentence 'A man in medieval costume walks through a rose garden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in medieval costume walks through a rose garden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 10.0], "base_se_norm": [-0.9621785283088684, -0.8108925819396973], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.9621785283088684, -0.9054462909698486], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000165.pt"}
{"idx": 167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-Z8xxygd2Y.mp4", "duration": 105.76, "prompt_text": "Please find the visual event described by the sentence 'The prince kisses the sleeping woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The prince kisses the sleeping woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [-0.01664149761199951, 0.07791221141815186], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.03555220365524292, 0.05900144577026367], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000167.pt"}
{"idx": 169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-Z8xxygd2Y.mp4", "duration": 105.76, "prompt_text": "Please find the visual event described by the sentence 'A man points a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man points a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 83.0], "base_se_norm": [0.4939485788345337, 0.5695915222167969], "gt_se_raw": [79.0, 82.0], "gt_se_norm": [0.4939485788345337, 0.5506807565689087], "start_token_span": [60, 64], "end_token_span": [66, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000169.pt"}
{"idx": 171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H7pK1f-mtoI.mp4", "duration": 69.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point do they give each other a high-five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they give each other a high-five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.34089547395706177, -0.25492537021636963], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.1402984857559204, -0.0829850435256958], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000171.pt"}
{"idx": 173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H7pK1f-mtoI.mp4", "duration": 69.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy open his mouth wide to laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy open his mouth wide to laugh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.28358209133148193, -0.19761192798614502], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.0829850435256958, -0.02567160129547119], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000173.pt"}
{"idx": 175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g2wsNw07wRY.mp4", "duration": 132.46566666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the spotted Great Dane first bend down to drink water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the spotted Great Dane first bend down to drink water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [-0.335676908493042, -0.2450873851776123], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.335676908493042, -0.27528393268585205], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000175.pt"}
{"idx": 177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g2wsNw07wRY.mp4", "duration": 132.46566666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a yellow dog jump into the pool from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a yellow dog jump into the pool from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 91.0], "base_se_norm": [0.3437443971633911, 0.37394094467163086], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [0.0870741605758667, 0.1172705888748169], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000177.pt"}
{"idx": 179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g2wsNw07wRY.mp4", "duration": 132.46566666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a black dog leaps into the pool making a big splash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a black dog leaps into the pool making a big splash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 83.0], "base_se_norm": [0.20786011219024658, 0.2531548738479614], "gt_se_raw": [80.0, 82.0], "gt_se_norm": [0.20786011219024658, 0.23805665969848633], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000179.pt"}
{"idx": 181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tT7jelfqCCw.mp4", "duration": 80.34693333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the nameplate for Dr. Kimberly McCorkle is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the nameplate for Dr. Kimberly McCorkle is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [-0.2781304717063904, -0.22834640741348267], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.2781304717063904, -0.20345431566238403], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000181.pt"}
{"idx": 183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tT7jelfqCCw.mp4", "duration": 80.34693333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the dates 'Friday, April 3rd' and 'Friday, May 1st' are mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the dates 'Friday, April 3rd' and 'Friday, May 1st' are mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.8755397200584412, -0.775971531867981], "gt_se_raw": [50.0, 54.0], "gt_se_norm": [0.24460256099700928, 0.34417080879211426], "start_token_span": [82, 85], "end_token_span": [87, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000183.pt"}
{"idx": 185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tODlPxhbGF8.mp4", "duration": 144.35, "prompt_text": "Please find the visual event described by the sentence 'When can we see the animated letter race on the giant screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the animated letter race on the giant screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 38.0], "base_se_norm": [-0.7783166170120239, -0.47350192070007324], "gt_se_raw": [11.0, 39.0], "gt_se_norm": [-0.8475926518440247, -0.45964670181274414], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000185.pt"}
{"idx": 187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tODlPxhbGF8.mp4", "duration": 144.35, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the message 'K Wins!' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the message 'K Wins!' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 39.0], "base_se_norm": [-0.5289227962493896, -0.45964670181274414], "gt_se_raw": [35.0, 39.0], "gt_se_norm": [-0.5150675773620605, -0.45964670181274414], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000187.pt"}
{"idx": 189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tODlPxhbGF8.mp4", "duration": 144.35, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when both teams are gathered in separate huddles on the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when both teams are gathered in separate huddles on the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.8753030896186829], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, -0.8753030896186829], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000189.pt"}
{"idx": 191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1TE0HIGNStc.mp4", "duration": 151.61813333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with the blue cap get into the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with the blue cap get into the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [-0.4723586440086365, -0.36683034896850586], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.4723586440086365, -0.4195944666862488], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000191.pt"}
{"idx": 193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1TE0HIGNStc.mp4", "duration": 151.61813333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman has a shocked expression after the man explains his condition.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman has a shocked expression after the man explains his condition.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 72.0], "base_se_norm": [-0.18215584754943848, -0.05024552345275879], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.06847381591796875, 0.10804688930511475], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000193.pt"}
{"idx": 195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bDird_PgnVs.mp4", "duration": 42.867000000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the woman hold the boy's hand and talk to him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman hold the boy's hand and talk to him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.7667202949523926], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.813376247882843], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000195.pt"}
{"idx": 197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bDird_PgnVs.mp4", "duration": 42.867000000000004, "prompt_text": "Please find the visual event described by the sentence 'When are the blue blocks being arranged on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the blue blocks being arranged on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 28.0], "base_se_norm": [-0.3001609444618225, 0.3063662052154541], "gt_se_raw": [16.0, 27.0], "gt_se_norm": [-0.25350505113601685, 0.2597101926803589], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000197.pt"}
{"idx": 199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bDird_PgnVs.mp4", "duration": 42.867000000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy pick up the red object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy pick up the red object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [0.7729254961013794, 0.9128932952880859], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [0.8195815086364746, 0.9128932952880859], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000199.pt"}
{"idx": 201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fZ9WiuJPnNA.mp4", "duration": 140.50703333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman is surrounded by four colorful monsters and they start singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman is surrounded by four colorful monsters and they start singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 39.0], "base_se_norm": [-0.7437850832939148, -0.44486767053604126], "gt_se_raw": [18.0, 27.0], "gt_se_norm": [-0.7437850832939148, -0.6156775951385498], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000201.pt"}
{"idx": 203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fZ9WiuJPnNA.mp4", "duration": 140.50703333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the woman appears to be flying or falling downwards while holding the number four.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the woman appears to be flying or falling downwards while holding the number four.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 70.0], "base_se_norm": [-0.16018438339233398, -0.0036085844039916992], "gt_se_raw": [62.0, 70.0], "gt_se_norm": [-0.11748188734054565, -0.0036085844039916992], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000203.pt"}
{"idx": 205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iYdb1mr7YhQ.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a man in a green shirt is seen from behind, sitting at a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a man in a green shirt is seen from behind, sitting at a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000205.pt"}
{"idx": 207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iYdb1mr7YhQ.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the teacher in a yellow dress writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the teacher in a yellow dress writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.04813218116760254, 0.3756735324859619], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000207.pt"}
{"idx": 209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iYdb1mr7YhQ.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'MEDICAL INFORMATION TECHNOLOGY' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'MEDICAL INFORMATION TECHNOLOGY' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.8689835071563721, -0.5414421558380127], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.8689835071563721, -0.5414421558380127], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000209.pt"}
{"idx": 211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I_D2UbSnkFg.mp4", "duration": 81.8818, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman dangles a beaded string in front of a boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman dangles a beaded string in front of a boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 73.0], "base_se_norm": [-0.047407329082489014, 0.7830580472946167], "gt_se_raw": [16.0, 58.0], "gt_se_norm": [-0.6091927289962769, 0.41667628288269043], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000211.pt"}
{"idx": 213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I_D2UbSnkFg.mp4", "duration": 81.8818, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman in the purple shirt is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman in the purple shirt is smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.1939600706100464, -0.09625822305679321], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [0.27012360095977783, 0.31897449493408203], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000213.pt"}
{"idx": 215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hgLQQe5uUCE.mp4", "duration": 89.75633333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the person walking along the sidewalk away from the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person walking along the sidewalk away from the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9777174592018127, -0.8217396140098572], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9777174592018127, -0.8440221548080444], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000215.pt"}
{"idx": 217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hgLQQe5uUCE.mp4", "duration": 89.75633333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame when the person is waiting at the corner by the stop sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame when the person is waiting at the corner by the stop sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 47.0], "base_se_norm": [-0.6880443096160889, 0.04727983474731445], "gt_se_raw": [36.0, 47.0], "gt_se_norm": [-0.19782817363739014, 0.04727983474731445], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000217.pt"}
{"idx": 219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hgLQQe5uUCE.mp4", "duration": 89.75633333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the person sitting in their seat on the bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the person sitting in their seat on the bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 81.0], "base_se_norm": [0.6266261339187622, 0.8048865795135498], "gt_se_raw": [72.0, 81.0], "gt_se_norm": [0.6043436527252197, 0.8048865795135498], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000219.pt"}
{"idx": 221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VpPJP7o7NnA.mp4", "duration": 68.56, "prompt_text": "Please find the visual event described by the sentence 'When is the list of 'Our Goals' being interacted with by a finger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the list of 'Our Goals' being interacted with by a finger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 44.0], "base_se_norm": [-0.18319714069366455, 0.2835472822189331], "gt_se_raw": [29.0, 44.0], "gt_se_norm": [-0.1540256142616272, 0.2835472822189331], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000221.pt"}
{"idx": 223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VpPJP7o7NnA.mp4", "duration": 68.56, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that displays the social media and website logos.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that displays the social media and website logos.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 54.0], "base_se_norm": [0.4002333879470825, 0.5752625465393066], "gt_se_raw": [47.0, 53.0], "gt_se_norm": [0.37106192111968994, 0.5460910797119141], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000223.pt"}
{"idx": 225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c8LraDxreeI.mp4", "duration": 129.56276666666668, "prompt_text": "Please find the visual event described by the sentence 'When are the gladiators fighting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the gladiators fighting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.7530154585838318, -0.6603962182998657], "gt_se_raw": [31.0, 37.0], "gt_se_norm": [-0.5214674472808838, -0.4288482069969177], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000225.pt"}
{"idx": 227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c8LraDxreeI.mp4", "duration": 129.56276666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man put a gladiator helmet on a child's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man put a gladiator helmet on a child's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.36710208654403687, -0.3207924962043762], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [-0.35166555643081665, -0.3207924962043762], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000227.pt"}
{"idx": 229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c8LraDxreeI.mp4", "duration": 129.56276666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a group of soldiers in formation holding their swords and shields?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a group of soldiers in formation holding their swords and shields?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.7530154585838318, -0.6603962182998657], "gt_se_raw": [44.0, 46.0], "gt_se_norm": [-0.3207924962043762, -0.2899194359779358], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000229.pt"}
{"idx": 231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uPUP8ey3MjE.mp4", "duration": 138.58333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an eagle swoops down and catches its prey in a field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an eagle swoops down and catches its prey in a field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [-0.2639807462692261, -0.22068548202514648], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.2639807462692261, -0.22068548202514648], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000231.pt"}
{"idx": 233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uPUP8ey3MjE.mp4", "duration": 138.58333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the eagle spreads its wings to shield the other bird in the nest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the eagle spreads its wings to shield the other bird in the nest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [-0.6392062306404114, -0.45159345865249634], "gt_se_raw": [27.0, 38.0], "gt_se_norm": [-0.6103427410125732, -0.45159345865249634], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000233.pt"}
{"idx": 235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TqPCGGHoxsE.mp4", "duration": 157.32383333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman hold up the blue and black shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman hold up the blue and black shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5550578832626343, -0.5169199705123901], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.5550578832626343, -0.5169199705123901], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000235.pt"}
{"idx": 237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TqPCGGHoxsE.mp4", "duration": 157.32383333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a loop is left in the middle of the shoelaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a loop is left in the middle of the shoelaces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 135.0], "base_se_norm": [0.5636537075042725, 0.7162052392959595], "gt_se_raw": [83.0, 87.0], "gt_se_norm": [0.05514848232269287, 0.10599899291992188], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000237.pt"}
{"idx": 239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TqPCGGHoxsE.mp4", "duration": 157.32383333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the presenter introduces herself and the topic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the presenter introduces herself and the topic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 39.0], "base_se_norm": [-0.9872873425483704, -0.5042073726654053], "gt_se_raw": [2.0, 12.0], "gt_se_norm": [-0.9745747447013855, -0.8474484086036682], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000239.pt"}
{"idx": 241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cbMupML0UIo.mp4", "duration": 129.23333333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the presenter is holding the book and flipping through the pages.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the presenter is holding the book and flipping through the pages.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 51.0], "base_se_norm": [-0.3809646964073181, -0.2107299566268921], "gt_se_raw": [38.0, 51.0], "gt_se_norm": [-0.41191643476486206, -0.2107299566268921], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000241.pt"}
{"idx": 243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cbMupML0UIo.mp4", "duration": 129.23333333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the phrase 'Disability Inclusion Starts With You' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the phrase 'Disability Inclusion Starts With You' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 128.0], "base_se_norm": [0.7023471593856812, 0.9809130430221558], "gt_se_raw": [110.0, 122.0], "gt_se_norm": [0.7023471593856812, 0.8880577087402344], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000243.pt"}
{"idx": 245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1AOXI0eBL8.mp4", "duration": 34.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man playing the guitar next to a fountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man playing the guitar next to a fountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.7678917050361633, -0.5938104391098022], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8259187936782837], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000245.pt"}
{"idx": 247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1AOXI0eBL8.mp4", "duration": 34.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time are people seen riding in a convertible car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are people seen riding in a convertible car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.3617021441459656, -0.24564796686172485], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.3036750555038452, -0.24564796686172485], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000247.pt"}
{"idx": 249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1AOXI0eBL8.mp4", "duration": 34.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does a man playfully cover a woman's eyes with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man playfully cover a woman's eyes with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 5.0], "base_se_norm": [-0.7678917050361633, -0.709864616394043], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [0.5087040662765503, 0.624758243560791], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000249.pt"}
{"idx": 251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lwS74rI92YQ.mp4", "duration": 23.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a plate is placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a plate is placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.6605374813079834, -0.4908062219619751], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.575671911239624, -0.4908062219619751], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000251.pt"}
{"idx": 253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lwS74rI92YQ.mp4", "duration": 23.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person put a mug on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person put a mug on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 22.0], "base_se_norm": [0.6973124742507935, 0.8670437335968018], "gt_se_raw": [21.0, 22.0], "gt_se_norm": [0.7821781635284424, 0.8670437335968018], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000253.pt"}
{"idx": 255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bf6D-i8YpHg.mp4", "duration": 189.08, "prompt_text": "Please find the visual event described by the sentence 'When does the Grinch get out of bed and smash the alarm clock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Grinch get out of bed and smash the alarm clock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.661518931388855, -0.6297863721847534], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.6509413719177246, -0.619208812713623], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000255.pt"}
{"idx": 257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bf6D-i8YpHg.mp4", "duration": 189.08, "prompt_text": "Please find the visual event described by the sentence 'At what point does the Grinch angrily throw a pickle jar into someone's shopping cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the Grinch angrily throw a pickle jar into someone's shopping cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 114.0], "base_se_norm": [0.15295112133026123, 0.20583879947662354], "gt_se_raw": [109.0, 112.0], "gt_se_norm": [0.15295112133026123, 0.1846836805343628], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000257.pt"}
{"idx": 259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bf6D-i8YpHg.mp4", "duration": 189.08, "prompt_text": "Please find the visual event described by the sentence 'When does the Grinch imagine himself gracefully ice skating?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Grinch imagine himself gracefully ice skating?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 153.0], "base_se_norm": [0.4173893928527832, 0.618362545967102], "gt_se_raw": [135.0, 142.0], "gt_se_norm": [0.4279669523239136, 0.5020097494125366], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000259.pt"}
{"idx": 261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0A8nEfYmmtM.mp4", "duration": 103.68691666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does one chip tell the other 'I don't think you know me'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one chip tell the other 'I don't think you know me'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 48.0], "base_se_norm": [-0.1127135157585144, -0.07413589954376221], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [-0.1127135157585144, -0.07413589954376221], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000261.pt"}
{"idx": 263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0A8nEfYmmtM.mp4", "duration": 103.68691666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part with the punchline 'I'm nacho friend'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part with the punchline 'I'm nacho friend'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 72.0], "base_se_norm": [0.1766190528869629, 0.38879621028900146], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [0.1959078311920166, 0.2537742853164673], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000263.pt"}
{"idx": 265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pCuE_A58ye0.mp4", "duration": 116.88, "prompt_text": "Please find the visual event described by the sentence 'When does Mr. Bean arrive at the cinema in his car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Mr. Bean arrive at the cinema in his car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8973305821418762, -0.8117727637290955], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.8973305821418762, -0.863107442855835], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000265.pt"}
{"idx": 267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pCuE_A58ye0.mp4", "duration": 116.88, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Mr. Bean is walking through the audience to find a seat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Mr. Bean is walking through the audience to find a seat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 44.0], "base_se_norm": [-0.36687201261520386, -0.24709099531173706], "gt_se_raw": [36.0, 44.0], "gt_se_norm": [-0.3839835524559021, -0.24709099531173706], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000267.pt"}
{"idx": 269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pCuE_A58ye0.mp4", "duration": 116.88, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the audience jumps in fright and Mr. Bean spills his popcorn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the audience jumps in fright and Mr. Bean spills his popcorn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 88.0], "base_se_norm": [0.4544832706451416, 0.5058180093765259], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.4544832706451416, 0.5058180093765259], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000269.pt"}
{"idx": 271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IYnsfV5N2n8.mp4", "duration": 107.4, "prompt_text": "Please find the visual event described by the sentence 'When does the evil tie come to life?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the evil tie come to life?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.6461825370788574, -0.5716946125030518], "gt_se_raw": [23.0, 24.0], "gt_se_norm": [-0.5716946125030518, -0.5530726313591003], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000271.pt"}
{"idx": 273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IYnsfV5N2n8.mp4", "duration": 107.4, "prompt_text": "Please find the visual event described by the sentence 'Show me when the potato gets squashed by a foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the potato gets squashed by a foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 65.0], "base_se_norm": [0.17318439483642578, 0.21042823791503906], "gt_se_raw": [64.0, 65.0], "gt_se_norm": [0.19180631637573242, 0.21042823791503906], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000273.pt"}
{"idx": 275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7QORMPFSzpc.mp4", "duration": 270.27, "prompt_text": "Please find the visual event described by the sentence 'When is the chicken being cut into pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the chicken being cut into pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.8963999152183533, -0.8593998551368713], "gt_se_raw": [13.0, 19.0], "gt_se_norm": [-0.9037998914718628, -0.8593998551368713], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000275.pt"}
{"idx": 277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7QORMPFSzpc.mp4", "duration": 270.27, "prompt_text": "Please find the visual event described by the sentence 'At what point are the diced vegetables mixed together in a colander?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the diced vegetables mixed together in a colander?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 96.0], "base_se_norm": [-0.31919926404953003, -0.2895992398262024], "gt_se_raw": [92.0, 96.0], "gt_se_norm": [-0.31919926404953003, -0.2895992398262024], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000277.pt"}
{"idx": 279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7QORMPFSzpc.mp4", "duration": 270.27, "prompt_text": "Please find the visual event described by the sentence 'When are two people seen assembling tacos at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 257.0 - 269.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two people seen assembling tacos at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 257.0 - 269.0 seconds.<|im_end|>\n", "base_se_raw": [257.0, 269.0], "base_se_norm": [0.9018019437789917, 0.9906020164489746], "gt_se_raw": [257.0, 269.0], "gt_se_norm": [0.9018019437789917, 0.9906020164489746], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000279.pt"}
{"idx": 281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmdAkgJDmEw.mp4", "duration": 194.31912499999999, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman in the yellow top appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman in the yellow top appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 96.0], "base_se_norm": [-0.5574290752410889, -0.01193457841873169], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.5574290752410889, -0.5368443727493286], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000281.pt"}
{"idx": 283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmdAkgJDmEw.mp4", "duration": 194.31912499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man picks up a large hammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man picks up a large hammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 107.0], "base_se_norm": [0.0601118803024292, 0.10128116607666016], "gt_se_raw": [102.0, 106.0], "gt_se_norm": [0.0498194694519043, 0.0909888744354248], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000283.pt"}
{"idx": 285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahj8Vef9L24.mp4", "duration": 381.13075, "prompt_text": "Please find the visual event described by the sentence 'When can we see the police convoy driving on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the police convoy driving on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [-0.5959391593933105, -0.5644539594650269], "gt_se_raw": [78.0, 83.0], "gt_se_norm": [-0.5906916260719299, -0.5644539594650269], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000285.pt"}
{"idx": 287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahj8Vef9L24.mp4", "duration": 381.13075, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the woman sees the bodies hanging from the overpass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the woman sees the bodies hanging from the overpass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 103.0], "base_se_norm": [-0.4857407808303833, -0.45950305461883545], "gt_se_raw": [98.0, 103.0], "gt_se_norm": [-0.4857407808303833, -0.45950305461883545], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000287.pt"}
{"idx": 289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahj8Vef9L24.mp4", "duration": 381.13075, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in sunglasses say 'Welcome to Juarez'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in sunglasses say 'Welcome to Juarez'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 139.0], "base_se_norm": [-0.30732429027557373, -0.27059149742126465], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [-0.5539588332176208, -0.5434637069702148], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000289.pt"}
{"idx": 291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIpev8JXJHQ.mp4", "duration": 175.50866666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the menu with only the word 'Malkovich' on it displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the menu with only the word 'Malkovich' on it displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 77.0], "base_se_norm": [-0.15673679113388062, -0.12255042791366577], "gt_se_raw": [74.0, 77.0], "gt_se_norm": [-0.15673679113388062, -0.12255042791366577], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000291.pt"}
{"idx": 293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIpev8JXJHQ.mp4", "duration": 175.50866666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man get thrown out of the restaurant through a window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get thrown out of the restaurant through a window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 119.0], "base_se_norm": [0.3218719959259033, 0.35605835914611816], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.3332674503326416, 0.36745381355285645], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000293.pt"}
{"idx": 295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nCLk51ELofU.mp4", "duration": 47.347, "prompt_text": "Please find the visual event described by the sentence 'When is the child looking at the green pom-pom while it is hanging in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the child looking at the green pom-pom while it is hanging in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, -0.28189748525619507], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, -0.15517348051071167], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000295.pt"}
{"idx": 297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nCLk51ELofU.mp4", "duration": 47.347, "prompt_text": "Please find the visual event described by the sentence 'When is the adult's hand shaking the tinsel around on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the adult's hand shaking the tinsel around on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 39.0], "base_se_norm": [0.09827446937561035, 0.6474117040634155], "gt_se_raw": [26.0, 34.0], "gt_se_norm": [0.09827446937561035, 0.43620502948760986], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000297.pt"}
{"idx": 299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nCLk51ELofU.mp4", "duration": 47.347, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the child turns their head to the left to look at the object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the child turns their head to the left to look at the object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.7887933850288391, -0.5775867700576782], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.36638015508651733, -0.23965615034103394], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000299.pt"}
{"idx": 301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5t7CB2N9oyU.mp4", "duration": 60.10170833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where we can see the 'Gotham Greens' logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where we can see the 'Gotham Greens' logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 39.0], "base_se_norm": [0.2312462329864502, 0.29780006408691406], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.2312462329864502, 0.29780006408691406], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000301.pt"}
{"idx": 303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5t7CB2N9oyU.mp4", "duration": 60.10170833333333, "prompt_text": "Please find the visual event described by the sentence 'A woman with blonde hair and a leopard print jacket smiles while talking to a man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman with blonde hair and a leopard print jacket smiles while talking to a man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.0016922354698181152, 0.06486153602600098], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.0016922354698181152, 0.06486153602600098], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000303.pt"}
{"idx": 305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tNeYTDLZUJA.mp4", "duration": 377.0099666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the initial rules explained, where even numbers and vowels are sorted to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the initial rules explained, where even numbers and vowels are sorted to the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 30.0], "base_se_norm": [-0.9734755158424377, -0.8408530354499817], "gt_se_raw": [4.0, 13.0], "gt_se_norm": [-0.9787803888320923, -0.9310362935066223], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000305.pt"}
{"idx": 307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tNeYTDLZUJA.mp4", "duration": 377.0099666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video explain the results of the study pairing faces with pleasant and unpleasant words?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 195.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video explain the results of the study pairing faces with pleasant and unpleasant words?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 195.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 195.0], "base_se_norm": [-0.06103283166885376, 0.03445541858673096], "gt_se_raw": [181.0, 195.0], "gt_se_norm": [-0.039813220500946045, 0.03445541858673096], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000307.pt"}
{"idx": 309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tNeYTDLZUJA.mp4", "duration": 377.0099666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the video introduce the hypothetical shooting task involving a person with an object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 253.0 - 278.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video introduce the hypothetical shooting task involving a person with an object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 253.0 - 278.0 seconds.<|im_end|>\n", "base_se_raw": [253.0, 278.0], "base_se_norm": [0.34213948249816895, 0.47476208209991455], "gt_se_raw": [253.0, 265.0], "gt_se_norm": [0.34213948249816895, 0.4057983160018921], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000309.pt"}
{"idx": 311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/chnJS5jGC5s.mp4", "duration": 123.15636666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman mentions the boy arching his back and crying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman mentions the boy arching his back and crying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.07434749603271484, 0.006850123405456543], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.2692216634750366, -0.23674267530441284], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000311.pt"}
{"idx": 313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/chnJS5jGC5s.mp4", "duration": 123.15636666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman demonstrate holding the handle of the bird ride?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman demonstrate holding the handle of the bird ride?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 108.0], "base_se_norm": [0.5752332210540771, 0.7538679838180542], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.5914727449417114, 0.6564308404922485], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000313.pt"}
{"idx": 315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MOQCYung8sA.mp4", "duration": 149.78296666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the man describe the accessibility of the ferry terminal with the Harbour Bridge in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man describe the accessibility of the ferry terminal with the Harbour Bridge in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 67.0], "base_se_norm": [-0.412483274936676, -0.10537230968475342], "gt_se_raw": [45.0, 68.0], "gt_se_norm": [-0.399130642414093, -0.09201961755752563], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000315.pt"}
{"idx": 317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MOQCYung8sA.mp4", "duration": 149.78296666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man explains the lack of accessible transport options while standing on a suburban street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man explains the lack of accessible transport options while standing on a suburban street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 100.0], "base_se_norm": [-0.051961660385131836, 0.33526527881622314], "gt_se_raw": [71.0, 101.0], "gt_se_norm": [-0.051961660385131836, 0.34861791133880615], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000317.pt"}
{"idx": 319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MOQCYung8sA.mp4", "duration": 149.78296666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker first introduce himself and the premise of the video while sitting on a couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker first introduce himself and the premise of the video while sitting on a couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 26.0], "base_se_norm": [-0.9065314531326294, -0.6528310775756836], "gt_se_raw": [7.0, 26.0], "gt_se_norm": [-0.9065314531326294, -0.6528310775756836], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000319.pt"}
{"idx": 321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5OLlemqd3SE.mp4", "duration": 70.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two people are present, and the one in the green shirt makes a swiping gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two people are present, and the one in the green shirt makes a swiping gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 25.0], "base_se_norm": [-0.46201038360595703, -0.2921189069747925], "gt_se_raw": [20.0, 24.0], "gt_se_norm": [-0.43369513750076294, -0.3204341530799866], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000321.pt"}
{"idx": 323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5OLlemqd3SE.mp4", "duration": 70.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a person wearing a bracelet is controlling the flight game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a person wearing a bracelet is controlling the flight game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 51.0], "base_se_norm": [-0.037281692028045654, 0.4440774917602539], "gt_se_raw": [35.0, 51.0], "gt_se_norm": [-0.008966445922851562, 0.4440774917602539], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000323.pt"}
{"idx": 325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdpspllWI2o.mp4", "duration": 92.0, "prompt_text": "Please find the visual event described by the sentence 'When does the orangutan swing on the houseplant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the orangutan swing on the houseplant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.804347813129425, -0.739130437374115], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.804347813129425, -0.739130437374115], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000325.pt"}
{"idx": 327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdpspllWI2o.mp4", "duration": 92.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the animation show the forest being destroyed by a large machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the animation show the forest being destroyed by a large machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.30434781312942505, -0.1304348111152649], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.239130437374115, -0.17391306161880493], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000327.pt"}
{"idx": 329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdpspllWI2o.mp4", "duration": 92.0, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the girl holds up the drawing that says 'SAVE MY FRIEND'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the girl holds up the drawing that says 'SAVE MY FRIEND'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 69.0], "base_se_norm": [0.43478262424468994, 0.5], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [0.43478262424468994, 0.5], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000329.pt"}
{"idx": 331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1roy4o4tqQM.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the man and Pikachu realize they can understand each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the man and Pikachu realize they can understand each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.19657063484191895, -0.11199915409088135], "gt_se_raw": [58.0, 64.0], "gt_se_norm": [-0.18247538805007935, -0.09790390729904175], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000331.pt"}
{"idx": 333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1roy4o4tqQM.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the man and Pikachu interrogating Mr. Mime in a dark room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the man and Pikachu interrogating Mr. Mime in a dark room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 127.0], "base_se_norm": [0.48000144958496094, 0.7900969982147217], "gt_se_raw": [106.0, 126.0], "gt_se_norm": [0.49409663677215576, 0.7760016918182373], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000333.pt"}
{"idx": 335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d61MkuYttDI.mp4", "duration": 192.05853333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the 'FROZEN' title appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the 'FROZEN' title appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.9166920781135559, -0.8333841562271118], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.9166920781135559, -0.8958650827407837], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000335.pt"}
{"idx": 337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d61MkuYttDI.mp4", "duration": 192.05853333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two princesses are on a swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two princesses are on a swing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [-0.7292492389678955, -0.6980087757110596], "gt_se_raw": [60.0, 61.0], "gt_se_norm": [-0.37519049644470215, -0.36477696895599365], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000337.pt"}
{"idx": 339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d61MkuYttDI.mp4", "duration": 192.05853333333334, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a display of various 'Frozen' merchandise like duct tape, butt wipes, and chicken soup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a display of various 'Frozen' merchandise like duct tape, butt wipes, and chicken soup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 134.0], "base_se_norm": [0.33292698860168457, 0.39540791511535645], "gt_se_raw": [128.0, 134.0], "gt_se_norm": [0.33292698860168457, 0.39540791511535645], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000339.pt"}
{"idx": 341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQXRmsvjwNU.mp4", "duration": 84.83474444444445, "prompt_text": "Please find the visual event described by the sentence 'At what point are the meatballs being fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the meatballs being fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.2455921173095703, -0.1512911319732666], "gt_se_raw": [31.0, 36.0], "gt_se_norm": [-0.26916736364364624, -0.1512911319732666], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000341.pt"}
{"idx": 343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQXRmsvjwNU.mp4", "duration": 84.83474444444445, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the pasta is added to the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the pasta is added to the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [0.20233750343322754, 0.29663848876953125], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [0.20233750343322754, 0.2494879961013794], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000343.pt"}
{"idx": 345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/84TouqfIsiI.mp4", "duration": 150.69220833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a man putting bags into the trunk of a black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a man putting bags into the trunk of a black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9601837396621704, -0.9070954322814941], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9601837396621704, -0.9070954322814941], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000345.pt"}
{"idx": 347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/84TouqfIsiI.mp4", "duration": 150.69220833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a yellow flower placed on a pillow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a yellow flower placed on a pillow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [-0.4425724148750305, -0.4160282015800476], "gt_se_raw": [42.0, 43.0], "gt_se_norm": [-0.4425724148750305, -0.42930030822753906], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000347.pt"}
{"idx": 349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/84TouqfIsiI.mp4", "duration": 150.69220833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a girl tries to use a walkie-talkie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a girl tries to use a walkie-talkie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 118.0], "base_se_norm": [0.5130177736282349, 0.5661060810089111], "gt_se_raw": [117.0, 118.0], "gt_se_norm": [0.5528340339660645, 0.5661060810089111], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000349.pt"}
{"idx": 351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oTuTdMMPfvc.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman in a floral dress introduces the dessert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman in a floral dress introduces the dessert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.5173316597938538, -0.4333893060684204], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.4753605127334595, -0.4333893060684204], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000351.pt"}
{"idx": 353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oTuTdMMPfvc.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a customer with unique sunglasses tastes the dessert and says 'It's delicious.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a customer with unique sunglasses tastes the dessert and says 'It's delicious.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [0.1751924753189087, 0.23814928531646729], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [0.1751924753189087, 0.23814928531646729], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000353.pt"}
{"idx": 355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24f0OwnZE-Q.mp4", "duration": 192.15029166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the plaid shirt walk into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the plaid shirt walk into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.9271403551101685, -0.8646892309188843], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9271403551101685, -0.9063233137130737], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000355.pt"}
{"idx": 357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24f0OwnZE-Q.mp4", "duration": 192.15029166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the man in the tie leans in to look at the other man's mustache.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the man in the tie leans in to look at the other man's mustache.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.48998260498046875, -0.45875704288482666], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.4795740842819214, -0.45875704288482666], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000357.pt"}
{"idx": 359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24f0OwnZE-Q.mp4", "duration": 192.15029166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man in the tie drops the hammer on the other man's foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man in the tie drops the hammer on the other man's foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 142.0], "base_se_norm": [0.43637561798095703, 0.4780097007751465], "gt_se_raw": [138.0, 140.0], "gt_se_norm": [0.43637561798095703, 0.45719265937805176], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000359.pt"}
{"idx": 361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gbicCoxSM1w.mp4", "duration": 66.2662, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a person is using a stapler to connect the papers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a person is using a stapler to connect the papers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 45.0], "base_se_norm": [-0.3360114097595215, 0.3581584692001343], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [0.14688944816589355, 0.2977958917617798], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000361.pt"}
{"idx": 363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gbicCoxSM1w.mp4", "duration": 66.2662, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the person crumples and shapes a small piece of paper with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the person crumples and shapes a small piece of paper with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.6378244161605835, -0.396373987197876], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [-0.577461838722229, -0.4265552759170532], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000363.pt"}
{"idx": 365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LbCBdiXpZNc.mp4", "duration": 88.95553333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman posing happily in front of a giant 'MELBOURNE' sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman posing happily in front of a giant 'MELBOURNE' sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.7077191472053528, -0.6402697563171387], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.7077191472053528, -0.6627528667449951], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000365.pt"}
{"idx": 367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LbCBdiXpZNc.mp4", "duration": 88.95553333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show four people gathered around a table for drinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show four people gathered around a table for drinks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 37.0], "base_se_norm": [-0.21309000253677368, -0.16812372207641602], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.21309000253677368, -0.16812372207641602], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000367.pt"}
{"idx": 369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LbCBdiXpZNc.mp4", "duration": 88.95553333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with the pillows that have a galaxy print on them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with the pillows that have a galaxy print on them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.32550573348999023, -0.28053945302963257], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.32550573348999023, -0.28053945302963257], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000369.pt"}
{"idx": 371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MYWVkW1u-o4.mp4", "duration": 148.3, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a group of four people are posing for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a group of four people are posing for a photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 33.0], "base_se_norm": [-0.6628456115722656, -0.5549561977386475], "gt_se_raw": [25.0, 32.0], "gt_se_norm": [-0.6628456115722656, -0.5684423446655273], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000371.pt"}
{"idx": 373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MYWVkW1u-o4.mp4", "duration": 148.3, "prompt_text": "Please find the visual event described by the sentence 'Find the clip showing a teacher instructing a student while looking at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip showing a teacher instructing a student while looking at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 96.0], "base_se_norm": [0.22724199295043945, 0.29467296600341797], "gt_se_raw": [91.0, 97.0], "gt_se_norm": [0.22724199295043945, 0.30815911293029785], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000373.pt"}
{"idx": 375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4EqgGiJkrpU.mp4", "duration": 203.63676666666666, "prompt_text": "Please find the visual event described by the sentence 'Please show me where the man details his main responsibilities at work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please show me where the man details his main responsibilities at work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 27.0], "base_se_norm": [-0.8723216652870178, -0.7348219156265259], "gt_se_raw": [16.0, 27.0], "gt_se_norm": [-0.8428574800491333, -0.7348219156265259], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000375.pt"}
{"idx": 377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4EqgGiJkrpU.mp4", "duration": 203.63676666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the interviewee share his favorite aspect of the job?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the interviewee share his favorite aspect of the job?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 73.0], "base_se_norm": [-0.45982247591018677, -0.28303712606430054], "gt_se_raw": [56.0, 73.0], "gt_se_norm": [-0.450001060962677, -0.28303712606430054], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000377.pt"}
{"idx": 379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4EqgGiJkrpU.mp4", "duration": 203.63676666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man discusses the application and interview experience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man discusses the application and interview experience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 125.0], "base_se_norm": [0.02142655849456787, 0.2276761531829834], "gt_se_raw": [104.0, 126.0], "gt_se_norm": [0.02142655849456787, 0.23749756813049316], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000379.pt"}
{"idx": 381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vYLaKMpqnOc.mp4", "duration": 67.3673, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a person in a white shirt and jeans walks from left to right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a person in a white shirt and jeans walks from left to right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8515600562095642, -0.7031201124191284], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.7624960541725159, -0.7031201124191284], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000381.pt"}
{"idx": 383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vYLaKMpqnOc.mp4", "duration": 67.3673, "prompt_text": "Please find the visual event described by the sentence 'When does the man adjust the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man adjust the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [0.7219035625457764, 0.8703435659408569], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.7515915632247925, 0.8406555652618408], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000383.pt"}
{"idx": 385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-PN1gwDOa5I.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the speaker shown wearing a crown and a detailed costume?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the speaker shown wearing a crown and a detailed costume?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 102.0], "base_se_norm": [-0.9628163576126099, 0.8963665962219238], "gt_se_raw": [2.0, 102.0], "gt_se_norm": [-0.9628163576126099, 0.8963665962219238], "start_token_span": [74, 77], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000385.pt"}
{"idx": 387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-PN1gwDOa5I.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where she describes an incident that happened during a performance of 'The King and I'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where she describes an incident that happened during a performance of 'The King and I'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 53.0], "base_se_norm": [-0.3121023178100586, -0.014632999897003174], "gt_se_raw": [49.0, 63.0], "gt_se_norm": [-0.08900034427642822, 0.17128527164459229], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000387.pt"}
{"idx": 389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elSiig58mOw.mp4", "duration": 156.78996666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman explain the function of the horizontal counting bar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman explain the function of the horizontal counting bar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 55.0], "base_se_norm": [-0.6045665144920349, -0.29842448234558105], "gt_se_raw": [62.0, 69.0], "gt_se_norm": [-0.20913302898406982, -0.11984163522720337], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000389.pt"}
{"idx": 391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elSiig58mOw.mp4", "duration": 156.78996666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the instructor explains the place values for the millions section.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the instructor explains the place values for the millions section.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 55.0], "base_se_norm": [-0.6045665144920349, -0.29842448234558105], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [0.1735445261001587, 0.21181225776672363], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000391.pt"}
{"idx": 393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elSiig58mOw.mp4", "duration": 156.78996666666666, "prompt_text": "Please find the visual event described by the sentence 'When is it explained that the bead above the counting bar has a value of five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is it explained that the bead above the counting bar has a value of five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 55.0], "base_se_norm": [-0.6045665144920349, -0.29842448234558105], "gt_se_raw": [48.0, 54.0], "gt_se_norm": [-0.3877159357070923, -0.3111804127693176], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000393.pt"}
{"idx": 395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6yd2tv4Ni4c.mp4", "duration": 88.2882, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the woman in the wheelchair starts driving away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the woman in the wheelchair starts driving away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 77.0], "base_se_norm": [0.6310219764709473, 0.7442874908447266], "gt_se_raw": [73.0, 77.0], "gt_se_norm": [0.6536750793457031, 0.7442874908447266], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000395.pt"}
{"idx": 397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6yd2tv4Ni4c.mp4", "duration": 88.2882, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman in the red shirt is talking about a Bible verse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman in the red shirt is talking about a Bible verse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 36.0], "base_se_norm": [-0.5469383001327515, -0.18448901176452637], "gt_se_raw": [51.0, 63.0], "gt_se_norm": [0.15530729293823242, 0.42714428901672363], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000397.pt"}
{"idx": 399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oY6tCnu-1Do.mp4", "duration": 85.1851, "prompt_text": "Please find the visual event described by the sentence 'A person defies gravity and floats away. When does this happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person defies gravity and floats away. When does this happen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.24869489669799805, -0.20173829793930054], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.24869489669799805, -0.20173829793930054], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000399.pt"}
{"idx": 401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oY6tCnu-1Do.mp4", "duration": 85.1851, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a baby is engulfed in flames.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a baby is engulfed in flames.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 62.0], "base_se_norm": [0.38521885871887207, 0.45565366744995117], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [0.38521885871887207, 0.4321753978729248], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000401.pt"}
{"idx": 403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oY6tCnu-1Do.mp4", "duration": 85.1851, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy who likes trains gets hit by one.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy who likes trains gets hit by one.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 79.0], "base_se_norm": [0.8078279495239258, 0.8547844886779785], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [0.8078279495239258, 0.8547844886779785], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000403.pt"}
{"idx": 405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBQzXWMb7wk.mp4", "duration": 121.05426666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the child is programming the robot by pressing the colored buttons on its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the child is programming the robot by pressing the colored buttons on its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 52.0], "base_se_norm": [-0.6034836173057556, -0.14088118076324463], "gt_se_raw": [27.0, 51.0], "gt_se_norm": [-0.5539190769195557, -0.15740269422531128], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000405.pt"}
{"idx": 407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBQzXWMb7wk.mp4", "duration": 121.05426666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the children are adjusting the placement of the orange tunnels on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the children are adjusting the placement of the orange tunnels on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 46.0], "base_se_norm": [-0.3721824288368225, -0.24001026153564453], "gt_se_raw": [38.0, 44.0], "gt_se_norm": [-0.3721824288368225, -0.27305328845977783], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000407.pt"}
{"idx": 409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x2-MCPa_3rU.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'When are the bullies chasing the boy on their bicycles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the bullies chasing the boy on their bicycles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 50.0], "base_se_norm": [-0.6535971760749817, -0.3814235329627991], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.6412256360054016, -0.6041110754013062], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000409.pt"}
{"idx": 411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x2-MCPa_3rU.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl in the pink dress help the boy up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl in the pink dress help the boy up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [-0.9133992791175842, -0.8886562585830688], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9133992791175842, -0.8886562585830688], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000411.pt"}
{"idx": 413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x2-MCPa_3rU.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'When does the boy run freely without his leg braces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy run freely without his leg braces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 120.0], "base_se_norm": [0.28663909435272217, 0.4845834970474243], "gt_se_raw": [78.0, 83.0], "gt_se_norm": [-0.03502070903778076, 0.02683699131011963], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000413.pt"}
{"idx": 415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/94NanQuVkA4.mp4", "duration": 100.01657777777778, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the onions are being cooked in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the onions are being cooked in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 43.0], "base_se_norm": [-0.3801027536392212, -0.140142560005188], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.3801027536392212, -0.30011600255966187], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000415.pt"}
{"idx": 417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/94NanQuVkA4.mp4", "duration": 100.01657777777778, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the peppers are being stuffed with all the ingredients.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the peppers are being stuffed with all the ingredients.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 58.0], "base_se_norm": [0.039827585220336914, 0.15980768203735352], "gt_se_raw": [53.0, 57.0], "gt_se_norm": [0.059824347496032715, 0.13981103897094727], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000417.pt"}
{"idx": 419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMBlu-8z_Mk.mp4", "duration": 142.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the stitches are cast onto the first needle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the stitches are cast onto the first needle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 34.0], "base_se_norm": [-0.8320895433425903, -0.5242537260055542], "gt_se_raw": [11.0, 27.0], "gt_se_norm": [-0.8460820913314819, -0.6222015023231506], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000419.pt"}
{"idx": 421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMBlu-8z_Mk.mp4", "duration": 142.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Where does the knitter arrange the needles into a triangle shape to join them in the round?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where does the knitter arrange the needles into a triangle shape to join them in the round?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 78.0], "base_se_norm": [-0.48227614164352417, 0.09141790866851807], "gt_se_raw": [68.0, 77.0], "gt_se_norm": [-0.0485074520111084, 0.07742536067962646], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000421.pt"}
{"idx": 423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMBlu-8z_Mk.mp4", "duration": 142.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'During which period is the person working on the second needle after joining in the round?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which period is the person working on the second needle after joining in the round?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 112.0], "base_se_norm": [0.17537307739257812, 0.5671641826629639], "gt_se_raw": [111.0, 118.0], "gt_se_norm": [0.5531716346740723, 0.6511193513870239], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000423.pt"}
{"idx": 425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7HDtvRHyq2M.mp4", "duration": 114.23912499999999, "prompt_text": "Please find the visual event described by the sentence 'At what time do we see the shot of graduates throwing confetti?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do we see the shot of graduates throwing confetti?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 92.0], "base_se_norm": [0.5581351518630981, 0.6106564998626709], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [0.5581351518630981, 0.6106564998626709], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000425.pt"}
{"idx": 427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7HDtvRHyq2M.mp4", "duration": 114.23912499999999, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a female soccer goalie makes a save.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a female soccer goalie makes a save.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 89.0], "base_se_norm": [0.5056136846542358, 0.5581351518630981], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [0.5056136846542358, 0.5581351518630981], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000427.pt"}
{"idx": 429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ddh13c7Hdo8.mp4", "duration": 11.04, "prompt_text": "Please find the visual event described by the sentence 'When is the FBI warning screen visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the FBI warning screen visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.8188405632972717, 0.8115942478179932], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8115942478179932], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000429.pt"}
{"idx": 431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ddh13c7Hdo8.mp4", "duration": 11.04, "prompt_text": "Please find the visual event described by the sentence 'At what time does the text 'a maximum penalty of up to five years in prison' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the text 'a maximum penalty of up to five years in prison' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.09420287609100342, 0.6304347515106201], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8115942478179932], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000431.pt"}
{"idx": 433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ddh13c7Hdo8.mp4", "duration": 11.04, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where 'FBI WARNING' is displayed in white text with a red underline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where 'FBI WARNING' is displayed in white text with a red underline.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.8188405632972717, 0.8115942478179932], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8115942478179932], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000433.pt"}
{"idx": 435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDquUlVDdbU.mp4", "duration": 178.278, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the person is handling the yellow watering can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the person is handling the yellow watering can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 63.0], "base_se_norm": [-0.4166414141654968, -0.29323863983154297], "gt_se_raw": [51.0, 59.0], "gt_se_norm": [-0.4278598427772522, -0.33811241388320923], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000435.pt"}
{"idx": 437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDquUlVDdbU.mp4", "duration": 178.278, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the person places the small plant pots on the windowsill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the person places the small plant pots on the windowsill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 134.0], "base_se_norm": [0.33499372005462646, 0.503270149230957], "gt_se_raw": [119.0, 134.0], "gt_se_norm": [0.33499372005462646, 0.503270149230957], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000437.pt"}
{"idx": 439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UNOKO0rFfgg.mp4", "duration": 141.72, "prompt_text": "Please find the visual event described by the sentence 'When can we see the shark swimming towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the shark swimming towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 18.0], "base_se_norm": [-0.9858876466751099, -0.7459779977798462], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.9858876466751099, -0.9294383525848389], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000439.pt"}
{"idx": 441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UNOKO0rFfgg.mp4", "duration": 141.72, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a large school of fish leaping out of the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a large school of fish leaping out of the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 57.0], "base_se_norm": [-0.25204628705978394, -0.19559693336486816], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.2379339337348938, -0.2097092866897583], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000441.pt"}
{"idx": 443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UNOKO0rFfgg.mp4", "duration": 141.72, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where sharks are hunting so close to the shore that their fins break the surface of the shallow water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where sharks are hunting so close to the shore that their fins break the surface of the shallow water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 116.0], "base_se_norm": [0.3265594244003296, 0.6370307207107544], "gt_se_raw": [96.0, 101.0], "gt_se_norm": [0.3547840118408203, 0.425345778465271], "start_token_span": [77, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000443.pt"}
{"idx": 445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9pH4c4fRgOg.mp4", "duration": 118.85206666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the data collection sheet for the triangle board shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the data collection sheet for the triangle board shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 112.0], "base_se_norm": [0.699591875076294, 0.8846958875656128], "gt_se_raw": [101.0, 106.0], "gt_se_norm": [0.699591875076294, 0.7837300300598145], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000445.pt"}
{"idx": 447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9pH4c4fRgOg.mp4", "duration": 118.85206666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man lying on his back on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man lying on his back on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.19227319955825806, -0.10813498497009277], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.19227319955825806, -0.1417902708053589], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000447.pt"}
{"idx": 449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z3biFxZIJOQ.mp4", "duration": 148.8153333333333, "prompt_text": "Please find the visual event described by the sentence 'When does Baymax attempt to fix a hole in himself using tape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Baymax attempt to fix a hole in himself using tape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [-0.27426832914352417, -0.16675257682800293], "gt_se_raw": [51.0, 70.0], "gt_se_norm": [-0.3145867586135864, -0.059236764907836914], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000449.pt"}
{"idx": 451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z3biFxZIJOQ.mp4", "duration": 148.8153333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is Hiro putting the red armor on Baymax?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is Hiro putting the red armor on Baymax?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 97.0], "base_se_norm": [0.26331067085266113, 0.3036290407180786], "gt_se_raw": [94.0, 98.0], "gt_se_norm": [0.26331067085266113, 0.31706857681274414], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000451.pt"}
{"idx": 453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z3biFxZIJOQ.mp4", "duration": 148.8153333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the car chase where Baymax is riding on the roof.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the car chase where Baymax is riding on the roof.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 110.0], "base_se_norm": [0.39770543575286865, 0.47834229469299316], "gt_se_raw": [104.0, 110.0], "gt_se_norm": [0.39770543575286865, 0.47834229469299316], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000453.pt"}
{"idx": 455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B9hkIBvBxOQ.mp4", "duration": 304.6043, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a brown bear is lying on a log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a brown bear is lying on a log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 62.0], "base_se_norm": [-0.6060463190078735, -0.5929144620895386], "gt_se_raw": [61.0, 62.0], "gt_se_norm": [-0.599480390548706, -0.5929144620895386], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000455.pt"}
{"idx": 457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B9hkIBvBxOQ.mp4", "duration": 304.6043, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where chunks of ice are breaking off the glacier and falling into the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 222.0 - 236.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where chunks of ice are breaking off the glacier and falling into the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 222.0 - 236.0 seconds.<|im_end|>\n", "base_se_raw": [222.0, 236.0], "base_se_norm": [0.4576287269592285, 0.549551248550415], "gt_se_raw": [223.0, 228.0], "gt_se_norm": [0.464194655418396, 0.49702417850494385], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000457.pt"}
{"idx": 459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bSEYSZLzEEo.mp4", "duration": 95.6956, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man smiling broadly at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man smiling broadly at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9373012185096741], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9373012185096741], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000459.pt"}
{"idx": 461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bSEYSZLzEEo.mp4", "duration": 95.6956, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman take the man's hand to help him make a choice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman take the man's hand to help him make a choice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 28.0], "base_se_norm": [-0.6238071918487549, -0.4148111343383789], "gt_se_raw": [20.0, 24.0], "gt_se_norm": [-0.5820080041885376, -0.49840956926345825], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000461.pt"}
{"idx": 463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bSEYSZLzEEo.mp4", "duration": 95.6956, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman is holding and stroking the man's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman is holding and stroking the man's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 36.0], "base_se_norm": [-0.5611083507537842, -0.247614324092865], "gt_se_raw": [58.0, 66.0], "gt_se_norm": [0.21217691898345947, 0.37937378883361816], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000463.pt"}
{"idx": 465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KLd5Q9IL_cU.mp4", "duration": 36.0, "prompt_text": "Please find the visual event described by the sentence 'At what moment is a reflection of a landscape visible in a pair of sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is a reflection of a landscape visible in a pair of sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.7222222089767456, -0.5555555820465088], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.7222222089767456, -0.5555555820465088], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000465.pt"}
{"idx": 467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KLd5Q9IL_cU.mp4", "duration": 36.0, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range showing a harbor with boats and a city skyline at dusk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range showing a harbor with boats and a city skyline at dusk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.11111116409301758, 0.2777777910232544], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.11111116409301758, 0.2777777910232544], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000467.pt"}
{"idx": 469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OTmPw4iy0hk.mp4", "duration": 166.4663, "prompt_text": "Please find the visual event described by the sentence 'When is Stephen Hawking shown sitting in his wheelchair at home?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Stephen Hawking shown sitting in his wheelchair at home?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 38.0], "base_se_norm": [-0.7356822490692139, -0.5434511303901672], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.7236677408218384, -0.6996389031410217], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000469.pt"}
{"idx": 471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OTmPw4iy0hk.mp4", "duration": 166.4663, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the female Intel Labs engineer is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the female Intel Labs engineer is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.31517666578292847, -0.24309003353118896], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [-0.30316221714019775, -0.2551044821739197], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000471.pt"}
{"idx": 473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OTmPw4iy0hk.mp4", "duration": 166.4663, "prompt_text": "Please find the visual event described by the sentence 'When are people seen gathered around large SGI computer servers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people seen gathered around large SGI computer servers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 146.0], "base_se_norm": [0.7180655002593994, 0.7541089057922363], "gt_se_raw": [143.0, 145.0], "gt_se_norm": [0.7180655002593994, 0.7420943975448608], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000473.pt"}
{"idx": 475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-vSmjbrUMpg.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the part that introduces the three sections of the guide.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part that introduces the three sections of the guide.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.692307710647583], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.38461536169052124, -0.29230767488479614], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000475.pt"}
{"idx": 477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-vSmjbrUMpg.mp4", "duration": 65.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the speaker talks about a list of reminders for Creative Commons licenses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the speaker talks about a list of reminders for Creative Commons licenses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 52.0], "base_se_norm": [0.015384674072265625, 0.6000000238418579], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [0.3538461923599243, 0.5692307949066162], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000477.pt"}
{"idx": 479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_0F7wFPRBmY.mp4", "duration": 188.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a black and white dog running on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a black and white dog running on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 90.0], "base_se_norm": [-0.10843801498413086, -0.044754981994628906], "gt_se_raw": [85.0, 89.0], "gt_se_norm": [-0.09782415628433228, -0.05536884069442749], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000479.pt"}
{"idx": 481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_0F7wFPRBmY.mp4", "duration": 188.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the dog catching a yellow tennis ball in mid-air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog catching a yellow tennis ball in mid-air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 110.0], "base_se_norm": [0.11445248126983643, 0.16752171516418457], "gt_se_raw": [105.0, 110.0], "gt_se_norm": [0.11445248126983643, 0.16752171516418457], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000481.pt"}
{"idx": 483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_0F7wFPRBmY.mp4", "duration": 188.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a dog run across a snowy or icy surface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a dog run across a snowy or icy surface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 136.0], "base_se_norm": [0.4222536087036133, 0.44348132610321045], "gt_se_raw": [134.0, 136.0], "gt_se_norm": [0.4222536087036133, 0.44348132610321045], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000483.pt"}
{"idx": 485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zi5ISjgijko.mp4", "duration": 76.81006666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person's hand touch the communication device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person's hand touch the communication device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 37.0], "base_se_norm": [-0.5313113927841187, -0.03658461570739746], "gt_se_raw": [18.0, 38.0], "gt_se_norm": [-0.5313113927841187, -0.010546326637268066], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000485.pt"}
{"idx": 487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zi5ISjgijko.mp4", "duration": 76.81006666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the photo of two people pointing at a sign outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the photo of two people pointing at a sign outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 76.0], "base_se_norm": [0.822677731513977, 0.9789073467254639], "gt_se_raw": [71.0, 76.0], "gt_se_norm": [0.8487160205841064, 0.9789073467254639], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000487.pt"}
{"idx": 489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QHVJvqC4LVE.mp4", "duration": 165.62379166666665, "prompt_text": "Please find the visual event described by the sentence 'When is the man squeezing toothpaste onto the toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man squeezing toothpaste onto the toothbrush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.43244868516921997, -0.323768675327301], "gt_se_raw": [48.0, 55.0], "gt_se_norm": [-0.4203731417655945, -0.3358442187309265], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000489.pt"}
{"idx": 491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QHVJvqC4LVE.mp4", "duration": 165.62379166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man rinse his mouth with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man rinse his mouth with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 93.0], "base_se_norm": [0.05057370662689209, 0.12302708625793457], "gt_se_raw": [87.0, 91.0], "gt_se_norm": [0.05057370662689209, 0.09887588024139404], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000491.pt"}
{"idx": 493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QHVJvqC4LVE.mp4", "duration": 165.62379166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man makes a disgusted face after finishing the juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man makes a disgusted face after finishing the juice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 140.0], "base_se_norm": [0.5577472448348999, 0.6905783414840698], "gt_se_raw": [129.0, 134.0], "gt_se_norm": [0.5577472448348999, 0.6181249618530273], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000493.pt"}
{"idx": 495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQwD0QCbxaA.mp4", "duration": 55.0, "prompt_text": "Please find the visual event described by the sentence 'Show me when the cat attempts to push the referee forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the cat attempts to push the referee forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.3454545736312866, -0.16363638639450073], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.018181800842285156, 0.09090912342071533], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000495.pt"}
{"idx": 497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQwD0QCbxaA.mp4", "duration": 55.0, "prompt_text": "Please find the visual event described by the sentence 'When do all the animals start to cheer and celebrate the win?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the animals start to cheer and celebrate the win?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 45.0], "base_se_norm": [0.30909085273742676, 0.6363636255264282], "gt_se_raw": [36.0, 44.0], "gt_se_norm": [0.30909085273742676, 0.6000000238418579], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000497.pt"}
{"idx": 499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vp01GKtd2AQ.mp4", "duration": 58.47508333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman's ice cream fall on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman's ice cream fall on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [-0.7605817914009094, -0.6921765804290771], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.7605817914009094, -0.6921765804290771], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000499.pt"}
{"idx": 501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vp01GKtd2AQ.mp4", "duration": 58.47508333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time do the two men look angry while on their phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do the two men look angry while on their phones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.35015052556991577, -0.28174537420272827], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.35015052556991577, -0.28174537420272827], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000501.pt"}
{"idx": 503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vp01GKtd2AQ.mp4", "duration": 58.47508333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the heart becomes happy after seeing the group of dogs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the heart becomes happy after seeing the group of dogs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 56.0], "base_se_norm": [0.8127379417419434, 0.9153457880020142], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [0.8127379417419434, 0.9153457880020142], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000503.pt"}
{"idx": 505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ri4xpRikZXg.mp4", "duration": 65.1651, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a boy puts on a gas mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a boy puts on a gas mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.6010134220123291, -0.4475570321083069], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.4782482981681824, -0.3861744999885559], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000505.pt"}
{"idx": 507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ri4xpRikZXg.mp4", "duration": 65.1651, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man in the red plaid shirt laughs triumphantly after the prank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man in the red plaid shirt laughs triumphantly after the prank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 34.0], "base_se_norm": [-0.29410070180892944, 0.043503403663635254], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [0.5038725137710571, 0.5652550458908081], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000507.pt"}
{"idx": 509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSz0zs0v6e8.mp4", "duration": 68.26820000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the blue character fall into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blue character fall into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.7070378065109253, -0.6191492080688477], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.6777416467666626, -0.6191492080688477], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000509.pt"}
{"idx": 511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSz0zs0v6e8.mp4", "duration": 68.26820000000001, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where two women in black are sitting together at a funeral.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where two women in black are sitting together at a funeral.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 44.0], "base_se_norm": [-0.2675946354866028, 0.28903353214263916], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.44337189197540283, -0.3261870741844177], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000511.pt"}
{"idx": 513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSz0zs0v6e8.mp4", "duration": 68.26820000000001, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up of the letter's content.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up of the letter's content.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 53.0], "base_se_norm": [0.31832969188690186, 0.5526994466781616], "gt_se_raw": [44.0, 64.0], "gt_se_norm": [0.28903353214263916, 0.874957799911499], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000513.pt"}
{"idx": 515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LSfJQkA-bKE.mp4", "duration": 81.91516666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of grilling the chicken breasts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of grilling the chicken breasts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 26.0], "base_se_norm": [-0.6337674856185913, -0.36519694328308105], "gt_se_raw": [16.0, 26.0], "gt_se_norm": [-0.6093519926071167, -0.36519694328308105], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000515.pt"}
{"idx": 517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LSfJQkA-bKE.mp4", "duration": 81.91516666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the crumbled goat cheese added and mixed into the salad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the crumbled goat cheese added and mixed into the salad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [0.4160991907119751, 0.5381766557693481], "gt_se_raw": [58.0, 63.0], "gt_se_norm": [0.4160991907119751, 0.5381766557693481], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000517.pt"}
{"idx": 519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpsh8dYl7PE.mp4", "duration": 40.33182707223309, "prompt_text": "Please find the visual event described by the sentence 'When can we see the animal from behind with a black hood over its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the animal from behind with a black hood over its head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 15.0], "base_se_norm": [-0.9504113793373108, -0.25617051124572754], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9504113793373108, -0.8016454577445984], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000519.pt"}
{"idx": 521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpsh8dYl7PE.mp4", "duration": 40.33182707223309, "prompt_text": "Please find the visual event described by the sentence 'Find the high-angle shot where the masked animal is on the left and the group of sheep is on the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the high-angle shot where the masked animal is on the left and the group of sheep is on the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 33.0], "base_se_norm": [0.19012713432312012, 0.6364247798919678], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [0.19012713432312012, 0.3388930559158325], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000521.pt"}
{"idx": 523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpsh8dYl7PE.mp4", "duration": 40.33182707223309, "prompt_text": "Please find the visual event described by the sentence 'When is the masked animal standing still and looking towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the masked animal standing still and looking towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.45452505350112915, -0.25617051124572754], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [0.7356021404266357, 0.8843679428100586], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000523.pt"}
{"idx": 525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FUoGS9Nx05M.mp4", "duration": 150.4, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person pushes a green button to open an automatic door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person pushes a green button to open an automatic door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.5744680762290955, -0.521276593208313], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.5744680762290955, -0.521276593208313], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000525.pt"}
{"idx": 527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FUoGS9Nx05M.mp4", "duration": 150.4, "prompt_text": "Please find the visual event described by the sentence 'When are the elevator buttons with braille numbers visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the elevator buttons with braille numbers visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [-0.13563823699951172, -0.09574466943740845], "gt_se_raw": [65.0, 69.0], "gt_se_norm": [-0.13563823699951172, -0.08244675397872925], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000527.pt"}
{"idx": 529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zXBmZLmfQZ4.mp4", "duration": 164.2, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in the white dress riding a bicycle downhill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in the white dress riding a bicycle downhill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [-0.5980511903762817, -0.512789249420166], "gt_se_raw": [33.0, 40.0], "gt_se_norm": [-0.5980511903762817, -0.512789249420166], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000529.pt"}
{"idx": 531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zXBmZLmfQZ4.mp4", "duration": 164.2, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man get thrown out of the van and land on the bicycle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man get thrown out of the van and land on the bicycle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.45188790559768677, -0.3666260242462158], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.3909865617752075, -0.3666260242462158], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000531.pt"}
{"idx": 533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zXBmZLmfQZ4.mp4", "duration": 164.2, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the dog takes control of the steering wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the dog takes control of the steering wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 44.0], "base_se_norm": [-0.512789249420166, -0.4640681743621826], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.5006090402603149, -0.47624844312667847], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000533.pt"}
{"idx": 535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/djqOMU5xpVo.mp4", "duration": 106.75055262710423, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the text 'Unfortunately this is all it kept doing on this stage...' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the text 'Unfortunately this is all it kept doing on this stage...' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.5878241062164307, -0.5316183567047119], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.5878241062164307, -0.5316183567047119], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000535.pt"}
{"idx": 537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/djqOMU5xpVo.mp4", "duration": 106.75055262710423, "prompt_text": "Please find the visual event described by the sentence 'Find the part where King Dedede hits Ness, sending him flying across the stage in a circular motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where King Dedede hits Ness, sending him flying across the stage in a circular motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 55.0], "base_se_norm": [-0.19438356161117554, 0.03043961524963379], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.1756482720375061, -0.10070723295211792], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000537.pt"}
{"idx": 539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9rxfM1pHQrY.mp4", "duration": 138.37170503837172, "prompt_text": "Please find the visual event described by the sentence 'When is a person ironing a red piece of clothing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person ironing a red piece of clothing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 69.0], "base_se_norm": [-0.10386300086975098, -0.0026862621307373047], "gt_se_raw": [62.0, 70.0], "gt_se_norm": [-0.10386300086975098, 0.01176750659942627], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000539.pt"}
{"idx": 541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9rxfM1pHQrY.mp4", "duration": 138.37170503837172, "prompt_text": "Please find the visual event described by the sentence 'At what point is a person flipping through a sketchbook with colored fashion illustrations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a person flipping through a sketchbook with colored fashion illustrations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 54.0], "base_se_norm": [-0.2773088812828064, -0.21949362754821777], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [-0.29176270961761475, -0.21949362754821777], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000541.pt"}
{"idx": 543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9rxfM1pHQrY.mp4", "duration": 138.37170503837172, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a woman with a pink pin cushion on her wrist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a woman with a pink pin cushion on her wrist?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.6097468137741089, -0.5519315004348755], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.6097468137741089, -0.5808391571044922], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000543.pt"}
{"idx": 545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UCuhC9jTFMM.mp4", "duration": 145.27012499999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a search bar with the text 'Make college affordable'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a search bar with the text 'Make college affordable'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5181390643119812, -0.4768366813659668], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.5181390643119812, -0.4768366813659668], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000545.pt"}
{"idx": 547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UCuhC9jTFMM.mp4", "duration": 145.27012499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is the phrase 'GO BLUE GUARANTEE' displayed on a yellow background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the phrase 'GO BLUE GUARANTEE' displayed on a yellow background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 64.0], "base_se_norm": [-0.17395269870758057, -0.11888283491134644], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.17395269870758057, -0.13265031576156616], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000547.pt"}
{"idx": 549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbgMz9omXik.mp4", "duration": 20.353666666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the child's hand towards the sippy cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the child's hand towards the sippy cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 9.0], "base_se_norm": [-0.8034752011299133, -0.11563843488693237], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.11563843488693237, 0.0808863639831543], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000549.pt"}
{"idx": 551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbgMz9omXik.mp4", "duration": 20.353666666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the child drink from the yellow and blue cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the child drink from the yellow and blue cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.017376065254211426], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [0.6704607009887695, 0.866985559463501], "start_token_span": [74, 77], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000551.pt"}
{"idx": 553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbgMz9omXik.mp4", "duration": 20.353666666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman stirs the food in the white bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman stirs the food in the white bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [0.0808863639831543, 0.3756735324859619], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.0808863639831543, 0.2774111032485962], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000553.pt"}
{"idx": 555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ObAw6Hv5n0k.mp4", "duration": 49.375, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the dog is standing on the bridge and looking at its reflection.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the dog is standing on the bridge and looking at its reflection.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 33.0], "base_se_norm": [-0.027848124504089355, 0.3367089033126831], "gt_se_raw": [24.0, 34.0], "gt_se_norm": [-0.027848124504089355, 0.3772151470184326], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000555.pt"}
{"idx": 557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ObAw6Hv5n0k.mp4", "duration": 49.375, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the dog is carrying the bone and walking towards home.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the dog is carrying the bone and walking towards home.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.351898729801178, -0.18987339735031128], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.351898729801178, -0.18987339735031128], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000557.pt"}
{"idx": 559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdkNEDjPes4.mp4", "duration": 147.31383333333332, "prompt_text": "Please find the visual event described by the sentence 'Show a shot of the Chicago city flag waving against a building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show a shot of the Chicago city flag waving against a building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 6.0], "base_se_norm": [-0.9321177005767822, -0.9185412526130676], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.9321177005767822, -0.9185412526130676], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000559.pt"}
{"idx": 561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdkNEDjPes4.mp4", "duration": 147.31383333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where an old woman with glasses and white hair is interviewed in a pink-walled room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where an old woman with glasses and white hair is interviewed in a pink-walled room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.4569416642189026, -0.29402416944503784], "gt_se_raw": [41.0, 52.0], "gt_se_norm": [-0.443365216255188, -0.29402416944503784], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000561.pt"}
{"idx": 563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JdkNEDjPes4.mp4", "duration": 147.31383333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the shot of a protester holding up a red sign that says 'BLACKLIVESMATTER'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot of a protester holding up a red sign that says 'BLACKLIVESMATTER'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 93.0], "base_se_norm": [0.2354576587677002, 0.2626105546951294], "gt_se_raw": [91.0, 93.0], "gt_se_norm": [0.2354576587677002, 0.2626105546951294], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000563.pt"}
{"idx": 565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5E_SFw2V99E.mp4", "duration": 76.31596548725838, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a basketball rolls and stops on the court.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a basketball rolls and stops on the court.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.18758803606033325, -0.10896754264831543], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.18758803606033325, -0.10896754264831543], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000565.pt"}
{"idx": 567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5E_SFw2V99E.mp4", "duration": 76.31596548725838, "prompt_text": "Please find the visual event described by the sentence 'At what point is a white canvas on an easel shown in a park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a white canvas on an easel shown in a park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 48.0], "base_se_norm": [0.1268939971923828, 0.25792813301086426], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [0.1268939971923828, 0.25792813301086426], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000567.pt"}
{"idx": 569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ofnCdC8P70g.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'When does Death touch the wolf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Death touch the wolf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [-0.8600000143051147, -0.8400000333786011], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.8600000143051147, -0.8400000333786011], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000569.pt"}
{"idx": 571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ofnCdC8P70g.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the deer nuzzle Death's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 214.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the deer nuzzle Death's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 214.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [214.0, 222.0], "base_se_norm": [0.42666661739349365, 0.48000001907348633], "gt_se_raw": [214.0, 222.0], "gt_se_norm": [0.42666661739349365, 0.48000001907348633], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000571.pt"}
{"idx": 573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ofnCdC8P70g.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when Death sits by the water next to a deer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 158.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when Death sits by the water next to a deer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 158.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [158.0, 163.0], "base_se_norm": [0.053333282470703125, 0.08666670322418213], "gt_se_raw": [157.0, 163.0], "gt_se_norm": [0.046666622161865234, 0.08666670322418213], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000573.pt"}
{"idx": 575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkjbMoj0JY4.mp4", "duration": 28.92, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man is holding three containers of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man is holding three containers of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 26.0], "base_se_norm": [0.03734433650970459, 0.7980636358261108], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.10096818208694458, 0.03734433650970459], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000575.pt"}
{"idx": 577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkjbMoj0JY4.mp4", "duration": 28.92, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man is collecting his fifth meal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man is collecting his fifth meal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 25.0], "base_se_norm": [0.31396961212158203, 0.7289073467254639], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.383125901222229, 0.521438479423523], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000577.pt"}
{"idx": 579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CaS0mXRHfGc.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person welding metal with sparks flying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person welding metal with sparks flying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.973060667514801, -0.9191819429397583], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.9461212754249573, -0.9191819429397583], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000579.pt"}
{"idx": 581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CaS0mXRHfGc.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man in the blue shirt introducing the lab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man in the blue shirt introducing the lab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 39.0], "base_se_norm": [-0.7755054235458374, -0.6497883796691895], "gt_se_raw": [25.0, 40.0], "gt_se_norm": [-0.7755054235458374, -0.640808641910553], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000581.pt"}
{"idx": 583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CaS0mXRHfGc.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the instructor is explaining something to two students next to a damaged silver SUV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the instructor is explaining something to two students next to a damaged silver SUV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 68.0], "base_se_norm": [-0.47019273042678833, -0.38937467336654663], "gt_se_raw": [59.0, 68.0], "gt_se_norm": [-0.47019273042678833, -0.38937467336654663], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000583.pt"}
{"idx": 585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HV-AmXJsVyc.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video show an animation of a chaotic crowd with flashing lights and loud noises?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show an animation of a chaotic crowd with flashing lights and loud noises?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [-0.41685670614242554, -0.2931596636772156], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.43452775478363037, -0.2754886746406555], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000585.pt"}
{"idx": 587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HV-AmXJsVyc.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'Sensitive Santa photography' website shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'Sensitive Santa photography' website shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 87.0], "base_se_norm": [0.3606675863265991, 0.5373777151107788], "gt_se_raw": [77.0, 82.0], "gt_se_norm": [0.3606675863265991, 0.44902265071868896], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000587.pt"}
{"idx": 589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b1XGPvbWn0A.mp4", "duration": 116.31620000000001, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman walking past a storefront with several mannequins in the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman walking past a storefront with several mannequins in the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.1574690341949463, -0.08869099617004395], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.5185537338256836, -0.44977569580078125], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000589.pt"}
{"idx": 591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b1XGPvbWn0A.mp4", "duration": 116.31620000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman walk past a couple who are looking at a map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman walk past a couple who are looking at a map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.6217207908630371, -0.5701372623443604], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.4153866767883301, -0.3809976577758789], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000591.pt"}
{"idx": 593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b1XGPvbWn0A.mp4", "duration": 116.31620000000001, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man wearing a white tank top and a red hat follows the woman through a construction walkway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man wearing a white tank top and a red hat follows the woman through a construction walkway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 83.0], "base_se_norm": [0.3583662509918213, 0.42714428901672363], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.3583662509918213, 0.42714428901672363], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000593.pt"}
{"idx": 595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y1jKP5m97Jk.mp4", "duration": 118.16666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man in the grey t-shirt folds his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man in the grey t-shirt folds his arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.5430183410644531, -0.49224257469177246], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.5430183410644531, -0.5091677904129028], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000595.pt"}
{"idx": 597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y1jKP5m97Jk.mp4", "duration": 118.16666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the grey shirt refer to a TV character?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the grey shirt refer to a TV character?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 88.0], "base_se_norm": [0.38787031173706055, 0.4894217252731323], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [0.4217207431793213, 0.47249650955200195], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000597.pt"}
{"idx": 599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ei6TjvfGMpE.mp4", "duration": 101.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the illustrations of disabled people shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the illustrations of disabled people shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 17.0], "base_se_norm": [-0.7835355997085571, -0.6654640436172485], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.7638570070266724, -0.6851426362991333], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000599.pt"}
{"idx": 601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ei6TjvfGMpE.mp4", "duration": 101.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the video present examples of slopes for wheelchair users?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the video present examples of slopes for wheelchair users?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 43.0], "base_se_norm": [-0.3702853322029114, -0.1538209319114685], "gt_se_raw": [33.0, 43.0], "gt_se_norm": [-0.3506067395210266, -0.1538209319114685], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000601.pt"}
{"idx": 603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ei6TjvfGMpE.mp4", "duration": 101.63333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the speaker presenting the conclusion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the speaker presenting the conclusion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 95.0], "base_se_norm": [0.16103637218475342, 0.8694654703140259], "gt_se_raw": [59.0, 95.0], "gt_se_norm": [0.16103637218475342, 0.8694654703140259], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000603.pt"}
{"idx": 605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M26ug8MGYlY.mp4", "duration": 139.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does one moose get knocked down onto the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one moose get knocked down onto the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.5126612782478333, -0.4553273320198059], "gt_se_raw": [35.0, 39.0], "gt_se_norm": [-0.4983277916908264, -0.4409938454627991], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000605.pt"}
{"idx": 607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M26ug8MGYlY.mp4", "duration": 139.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the moose are fighting near the stop sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the moose are fighting near the stop sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 117.0], "base_se_norm": [0.44768261909484863, 0.6770185232162476], "gt_se_raw": [100.0, 108.0], "gt_se_norm": [0.4333491325378418, 0.548017144203186], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000607.pt"}
{"idx": 609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z2BgjH_CtIA.mp4", "duration": 23.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the dog lying still on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog lying still on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, 0.4676259756088257], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.3956834077835083], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000609.pt"}
{"idx": 611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z2BgjH_CtIA.mp4", "duration": 23.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog suddenly wake up and crash into the object against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog suddenly wake up and crash into the object against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [0.4676259756088257, 0.726618766784668], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.5539568662643433, 0.726618766784668], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000611.pt"}
{"idx": 613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z2BgjH_CtIA.mp4", "duration": 23.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the dog rolls onto its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the dog rolls onto its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.38129496574401855, 0.5539568662643433], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.3093525171279907, -0.05035966634750366], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000613.pt"}
{"idx": 615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DCWHh_mZogE.mp4", "duration": 313.84, "prompt_text": "Please find the visual event described by the sentence 'At what point does Grandpa Pig give Peppa and George flower seeds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Grandpa Pig give Peppa and George flower seeds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.674993634223938, -0.6495029330253601], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.674993634223938, -0.655875563621521], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000615.pt"}
{"idx": 617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DCWHh_mZogE.mp4", "duration": 313.84, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the snail is eating the baby plants.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 213.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the snail is eating the baby plants.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 213.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 213.0], "base_se_norm": [0.30002546310424805, 0.3573795557022095], "gt_se_raw": [205.0, 209.0], "gt_se_norm": [0.3063981533050537, 0.33188891410827637], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000617.pt"}
{"idx": 619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8PWlxfTFmyE.mp4", "duration": 50.2, "prompt_text": "Please find the visual event described by the sentence 'When does the character exclaim, 'You've killed us all!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character exclaim, 'You've killed us all!'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.8406374454498291, -0.6812748908996582], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.7609561681747437, -0.7211155295372009], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000619.pt"}
{"idx": 621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8PWlxfTFmyE.mp4", "duration": 50.2, "prompt_text": "Please find the visual event described by the sentence 'At what time does the fax machine transform into a robot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the fax machine transform into a robot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.00398409366607666, 0.11553788185119629], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.00398409366607666, 0.11553788185119629], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000621.pt"}
{"idx": 623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8PWlxfTFmyE.mp4", "duration": 50.2, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man begins to drown in a puddle of milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man begins to drown in a puddle of milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.7211155295372009, -0.5617530345916748], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.7211155295372009, -0.6015936136245728], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000623.pt"}
{"idx": 625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RvzFTHdMPWw.mp4", "duration": 40.04004004004004, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the speaker explains why estimates are free.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the speaker explains why estimates are free.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 18.0], "base_se_norm": [-0.500499963760376, -0.10089999437332153], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.6003999710083008, -0.35065001249313354], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000625.pt"}
{"idx": 627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RvzFTHdMPWw.mp4", "duration": 40.04004004004004, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the speaker's name, Jean Cantave, is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the speaker's name, Jean Cantave, is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, 0.19880008697509766], "gt_se_raw": [0.0, 24.0], "gt_se_norm": [-1.0, 0.19880008697509766], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000627.pt"}
{"idx": 629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-AXQVPtK-E.mp4", "duration": 188.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see two players building bridges to get close to each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two players building bridges to get close to each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 139.0], "base_se_norm": [0.350824236869812, 0.47846126556396484], "gt_se_raw": [129.0, 131.0], "gt_se_norm": [0.3720971345901489, 0.3933699131011963], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000629.pt"}
{"idx": 631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-AXQVPtK-E.mp4", "duration": 188.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player in the yellow shirt get eliminated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in the yellow shirt get eliminated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 108.0], "base_se_norm": [0.11682319641113281, 0.14873242378234863], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [-0.3086332678794861, -0.28736042976379395], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000631.pt"}
{"idx": 633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-AXQVPtK-E.mp4", "duration": 188.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Please show me the moment when a player falls into lava.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please show me the moment when a player falls into lava.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.22318732738494873, 0.25509655475616455], "gt_se_raw": [115.0, 117.0], "gt_se_norm": [0.22318732738494873, 0.2444601058959961], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000633.pt"}
{"idx": 635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yad0gj-ZDuE.mp4", "duration": 119.9, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the green paper is folded into a small square shape with four flaps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the green paper is folded into a small square shape with four flaps.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 47.0], "base_se_norm": [-0.7497915029525757, -0.2160133719444275], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.2827356457710266, -0.24937450885772705], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000635.pt"}
{"idx": 637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yad0gj-ZDuE.mp4", "duration": 119.9, "prompt_text": "Please find the visual event described by the sentence 'At what time is the finished origami star being transformed back and forth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the finished origami star being transformed back and forth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 111.0], "base_se_norm": [0.567973256111145, 0.8515429496765137], "gt_se_raw": [94.0, 112.0], "gt_se_norm": [0.567973256111145, 0.8682235479354858], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000637.pt"}
{"idx": 639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NooW_RbfdWI.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a dinosaur's claw is seen approaching the girl's bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a dinosaur's claw is seen approaching the girl's bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.8321153521537781, -0.7481729984283447], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.8321153521537781, -0.7691586017608643], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000639.pt"}
{"idx": 641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NooW_RbfdWI.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the large dinosaur breaks out of the tunnel into the fiery room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the large dinosaur breaks out of the tunnel into the fiery room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.20254796743392944, -0.13959121704101562], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.1815623641014099, -0.13959121704101562], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000641.pt"}
{"idx": 643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NooW_RbfdWI.mp4", "duration": 95.30353333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man get trapped in a cage with a large dinosaur trying to bite him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man get trapped in a cage with a large dinosaur trying to bite him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 78.0], "base_se_norm": [0.5949040651321411, 0.6368752717971802], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [0.2801203727722168, 0.32209157943725586], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000643.pt"}
{"idx": 645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pq6qrbKiDIk.mp4", "duration": 154.22073333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man explains that people are a source of power for the military.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man explains that people are a source of power for the military.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 113.0], "base_se_norm": [0.3227792978286743, 0.4654320478439331], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [-0.2478313446044922, -0.1570523977279663], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000645.pt"}
{"idx": 647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pq6qrbKiDIk.mp4", "duration": 154.22073333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the speaker talks about people leaving their state to go to another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the speaker talks about people leaving their state to go to another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 101.0], "base_se_norm": [0.2190319299697876, 0.3098108768463135], "gt_se_raw": [93.0, 96.0], "gt_se_norm": [0.20606350898742676, 0.24496877193450928], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000647.pt"}
{"idx": 649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U5oHhI_GmJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman sitting on a stool as a model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman sitting on a stool as a model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.6069504022598267], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000649.pt"}
{"idx": 651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U5oHhI_GmJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man with glasses and a blue jacket looking at a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man with glasses and a blue jacket looking at a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08288431167602539, 0.3756735324859619], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000651.pt"}
{"idx": 653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U5oHhI_GmJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where the text 'YOU'VE GOT THIS!' is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where the text 'YOU'VE GOT THIS!' is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.5066900253295898, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000653.pt"}
{"idx": 655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DOyum_nw7NU.mp4", "duration": 220.95406666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the award for '2006 Best Teacher' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the award for '2006 Best Teacher' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 64.0], "base_se_norm": [-0.4478490352630615, -0.42069411277770996], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.4478490352630615, -0.4116424322128296], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000655.pt"}
{"idx": 657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DOyum_nw7NU.mp4", "duration": 220.95406666666668, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when Nathalie Kees is shown wearing a graduation gown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when Nathalie Kees is shown wearing a graduation gown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 49.0], "base_se_norm": [-0.5836238861083984, -0.5564689040184021], "gt_se_raw": [46.0, 49.0], "gt_se_norm": [-0.5836238861083984, -0.5564689040184021], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000657.pt"}
{"idx": 659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5yGNbyAmkVY.mp4", "duration": 18.666666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'CAUTION WATCH YOUR STEP' sign clearly visible and in focus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'CAUTION WATCH YOUR STEP' sign clearly visible and in focus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.5714285373687744, -0.3571428060531616], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.5714285373687744, -0.3571428060531616], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000659.pt"}
{"idx": 661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5yGNbyAmkVY.mp4", "duration": 18.666666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the character fall into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character fall into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.17857146263122559, 0.3928571939468384], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [0.17857146263122559, 0.28571438789367676], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000661.pt"}
{"idx": 663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5yGNbyAmkVY.mp4", "duration": 18.666666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does water splash up from the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does water splash up from the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [0.3928571939468384, 0.6071429252624512], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.3928571939468384, 0.6071429252624512], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000663.pt"}
{"idx": 665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jwPc0kK9VHU.mp4", "duration": 123.5, "prompt_text": "Please find the visual event described by the sentence 'When does the man toss an apple into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man toss an apple into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.38461536169052124, -0.3360323905944824], "gt_se_raw": [38.0, 39.0], "gt_se_norm": [-0.38461536169052124, -0.3684210777282715], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000665.pt"}
{"idx": 667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jwPc0kK9VHU.mp4", "duration": 123.5, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is dozing off at his apple pie stall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is dozing off at his apple pie stall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 92.0], "base_se_norm": [0.26315784454345703, 0.4898785352706909], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [0.3441295623779297, 0.4251011610031128], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000667.pt"}
{"idx": 669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8zkVKHy1hyM.mp4", "duration": 115.0732888888889, "prompt_text": "Please find the visual event described by the sentence 'When is yeast added to the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is yeast added to the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.9478592872619629, -0.9130988717079163], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9652395248413086, -0.930479109287262], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000669.pt"}
{"idx": 671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8zkVKHy1hyM.mp4", "duration": 115.0732888888889, "prompt_text": "Please find the visual event described by the sentence 'At what point is the risen dough placed onto the floured surface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the risen dough placed onto the floured surface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.14836883544921875, -0.07884788513183594], "gt_se_raw": [49.0, 53.0], "gt_se_norm": [-0.14836883544921875, -0.07884788513183594], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000671.pt"}
{"idx": 673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8zkVKHy1hyM.mp4", "duration": 115.0732888888889, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the pastry is being fried in hot oil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the pastry is being fried in hot oil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 87.0], "base_se_norm": [0.39041829109191895, 0.5120798349380493], "gt_se_raw": [81.0, 86.0], "gt_se_norm": [0.40779852867126465, 0.4946995973587036], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000673.pt"}
{"idx": 675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/60kjpwyQhqg.mp4", "duration": 15.708333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the character puts its hands on its hips.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the character puts its hands on its hips.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [0.018567681312561035, 0.2732095718383789], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [0.018567681312561035, 0.1458885669708252], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000675.pt"}
{"idx": 677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/60kjpwyQhqg.mp4", "duration": 15.708333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the character's eyes are closed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the character's eyes are closed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [0.1458885669708252, 0.9098143577575684], "gt_se_raw": [8.0, 15.0], "gt_se_norm": [0.018567681312561035, 0.9098143577575684], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000677.pt"}
{"idx": 679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EXJ4Dbdm1ks.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'When does the video show examples of intersections, such as being gay and black?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show examples of intersections, such as being gay and black?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 37.0], "base_se_norm": [-0.6803196668624878, -0.6184461116790771], "gt_se_raw": [31.0, 37.0], "gt_se_norm": [-0.6803196668624878, -0.6184461116790771], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000679.pt"}
{"idx": 681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EXJ4Dbdm1ks.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker explain who coined the term 'intersectionality'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker explain who coined the term 'intersectionality'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [-0.443137526512146, -0.3812639117240906], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.443137526512146, -0.4018884301185608], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000681.pt"}
{"idx": 683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EXJ4Dbdm1ks.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the speaker lists the different types of boxes people check off, like gender, sexuality, and race.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the speaker lists the different types of boxes people check off, like gender, sexuality, and race.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 48.0], "base_se_norm": [-0.5668847560882568, -0.5050111413002014], "gt_se_raw": [42.0, 48.0], "gt_se_norm": [-0.5668847560882568, -0.5050111413002014], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000683.pt"}
{"idx": 685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkbsMObgMBE.mp4", "duration": 127.96, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the mother leopard is carrying her cub.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the mother leopard is carrying her cub.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 115.0], "base_se_norm": [0.6098780632019043, 0.7974367141723633], "gt_se_raw": [104.0, 116.0], "gt_se_norm": [0.6255079507827759, 0.8130666017532349], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000685.pt"}
{"idx": 687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkbsMObgMBE.mp4", "duration": 127.96, "prompt_text": "Please find the visual event described by the sentence 'At what time does the leopard cub attempt to catch the rabbit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the leopard cub attempt to catch the rabbit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 80.0], "base_se_norm": [0.09409189224243164, 0.25039076805114746], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [0.09409189224243164, 0.1253516674041748], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000687.pt"}
{"idx": 689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vCTNioSW730.mp4", "duration": 174.44, "prompt_text": "Please find the visual event described by the sentence 'When does the squirrel try to bury the acorn in the snow, causing a massive crack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the squirrel try to bury the acorn in the snow, causing a massive crack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [-0.3350149393081665, -0.2776886224746704], "gt_se_raw": [54.0, 63.0], "gt_se_norm": [-0.38087594509124756, -0.2776886224746704], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000689.pt"}
{"idx": 691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vCTNioSW730.mp4", "duration": 174.44, "prompt_text": "Please find the visual event described by the sentence 'At what point does the squirrel start sliding rapidly through the narrow ice canyon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the squirrel start sliding rapidly through the narrow ice canyon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 142.0], "base_se_norm": [0.5248795747756958, 0.628066897392273], "gt_se_raw": [134.0, 142.0], "gt_se_norm": [0.5363448858261108, 0.628066897392273], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000691.pt"}
{"idx": 693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vCTNioSW730.mp4", "duration": 174.44, "prompt_text": "Please find the visual event described by the sentence 'Show when the squirrel finally reunites with the acorn and hugs it on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show when the squirrel finally reunites with the acorn and hugs it on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 172.0], "base_se_norm": [0.9146984815597534, 0.97202467918396], "gt_se_raw": [167.0, 172.0], "gt_se_norm": [0.9146984815597534, 0.97202467918396], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000693.pt"}
{"idx": 695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fPh8BTdzXzg.mp4", "duration": 309.9763333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the frozen mango mixture is cut into small cubes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the frozen mango mixture is cut into small cubes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 201.0], "base_se_norm": [0.15492689609527588, 0.29687321186065674], "gt_se_raw": [180.0, 199.0], "gt_se_norm": [0.16137909889221191, 0.28396904468536377], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000695.pt"}
{"idx": 697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fPh8BTdzXzg.mp4", "duration": 309.9763333333333, "prompt_text": "Please find the visual event described by the sentence 'Can you locate the part where the person is playing a game on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you locate the part where the person is playing a game on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 158.0], "base_se_norm": [-0.006375730037689209, 0.019432663917541504], "gt_se_raw": [155.0, 158.0], "gt_se_norm": [7.641315460205078e-05, 0.019432663917541504], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000697.pt"}
{"idx": 699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GbycvPwr1Wg.mp4", "duration": 147.93945833333333, "prompt_text": "Please find the visual event described by the sentence 'A man is shown with the word 'DOG' on his T-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is shown with the word 'DOG' on his T-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.6485048532485962, -0.6214667558670044], "gt_se_raw": [27.0, 28.0], "gt_se_norm": [-0.6349858045578003, -0.6214667558670044], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000699.pt"}
{"idx": 701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GbycvPwr1Wg.mp4", "duration": 147.93945833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when one man is woken up by another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when one man is woken up by another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 98.0], "base_se_norm": [0.14911878108978271, 0.32486629486083984], "gt_se_raw": [86.0, 95.0], "gt_se_norm": [0.1626378297805786, 0.28430914878845215], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000701.pt"}
{"idx": 703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GbycvPwr1Wg.mp4", "duration": 147.93945833333333, "prompt_text": "Please find the visual event described by the sentence 'Two men are talking in the kitchen while one of them is cooking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Two men are talking in the kitchen while one of them is cooking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 64.0], "base_se_norm": [-0.351085901260376, -0.13478118181228638], "gt_se_raw": [48.0, 56.0], "gt_se_norm": [-0.351085901260376, -0.2429335117340088], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000703.pt"}
{"idx": 705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y1TceAcocv0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker use the phrase 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker use the phrase 'nearly half the rate'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.03166300058364868, 0.16200435161590576], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.06517064571380615, 0.25883805751800537], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000705.pt"}
{"idx": 707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y1TceAcocv0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the speaker refers to 'those without disabilities' for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the speaker refers to 'those without disabilities' for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [0.25883805751800537, 0.5493391752243042], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [0.25883805751800537, 0.355671763420105], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000707.pt"}
{"idx": 709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T_FdjXqSZlc.mp4", "duration": 266.89996666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the girl gives the toy rocket back to her younger brother.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the girl gives the toy rocket back to her younger brother.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 95.0], "base_se_norm": [-0.33308345079421997, -0.2881227731704712], "gt_se_raw": [90.0, 94.0], "gt_se_norm": [-0.32559001445770264, -0.2956162095069885], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000709.pt"}
{"idx": 711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T_FdjXqSZlc.mp4", "duration": 266.89996666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the father start running back towards his son on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 243.0 - 255.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the father start running back towards his son on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 243.0 - 255.0 seconds.<|im_end|>\n", "base_se_raw": [243.0, 255.0], "base_se_norm": [0.8209069967269897, 0.9108282327651978], "gt_se_raw": [243.0, 255.0], "gt_se_norm": [0.8209069967269897, 0.9108282327651978], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000711.pt"}
{"idx": 713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C9O28ne6bG8.mp4", "duration": 111.3112, "prompt_text": "Please find the visual event described by the sentence 'Show me when a cat with long hair cautiously touches the lobster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a cat with long hair cautiously touches the lobster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8203235864639282, -0.7664206027984619], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.8203235864639282, -0.7843883037567139], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000713.pt"}
{"idx": 715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C9O28ne6bG8.mp4", "duration": 111.3112, "prompt_text": "Please find the visual event described by the sentence 'At what point do four cats gather around the green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do four cats gather around the green bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 38.0], "base_se_norm": [-0.5867441892623901, -0.31722956895828247], "gt_se_raw": [28.0, 35.0], "gt_se_norm": [-0.49690598249435425, -0.371132493019104], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000715.pt"}
{"idx": 717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C9O28ne6bG8.mp4", "duration": 111.3112, "prompt_text": "Please find the visual event described by the sentence 'Locate the close-up shot of the lobster and fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the close-up shot of the lobster and fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.31722956895828247, -0.24535900354385376], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.2992619276046753, -0.26332664489746094], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000717.pt"}
{"idx": 719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tbBzXKN32Sk.mp4", "duration": 472.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man start pouring liquid into the first glass on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man start pouring liquid into the first glass on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 65.0], "base_se_norm": [-0.7713479399681091, -0.7247706651687622], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [-0.7459421157836914, -0.7247706651687622], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000719.pt"}
{"idx": 721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tbBzXKN32Sk.mp4", "duration": 472.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man start drinking from the first glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 206.0 - 224.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start drinking from the first glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 206.0 - 224.0 seconds.<|im_end|>\n", "base_se_raw": [206.0, 224.0], "base_se_norm": [-0.12773466110229492, -0.0515173077583313], "gt_se_raw": [206.0, 224.0], "gt_se_norm": [-0.12773466110229492, -0.0515173077583313], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000721.pt"}
{"idx": 723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/adzYW5DZoWs.mp4", "duration": 123.83204166666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a beige outfit standing in the desert?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a beige outfit standing in the desert?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 24.0], "base_se_norm": [-0.838490903377533, -0.6123781800270081], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.838490903377533, -0.7900381684303284], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000723.pt"}
{"idx": 725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/adzYW5DZoWs.mp4", "duration": 123.83204166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a spacecraft exploding in the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a spacecraft exploding in the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 90.0], "base_se_norm": [0.405129075050354, 0.4535818099975586], "gt_se_raw": [87.0, 89.0], "gt_se_norm": [0.405129075050354, 0.43743085861206055], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000725.pt"}
{"idx": 727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/adzYW5DZoWs.mp4", "duration": 123.83204166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the scene with a large wrecked ship in a stormy sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene with a large wrecked ship in a stormy sea.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 104.0], "base_se_norm": [0.5666382312774658, 0.6796945333480835], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.5827890634536743, 0.647392749786377], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000727.pt"}
{"idx": 729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YzN4Zj9S6so.mp4", "duration": 157.62413333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a speaker introduces himself and the topic of his presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a speaker introduces himself and the topic of his presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 59.0], "base_se_norm": [-0.581282377243042, -0.2513836622238159], "gt_se_raw": [32.0, 59.0], "gt_se_norm": [-0.593970775604248, -0.2513836622238159], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000729.pt"}
{"idx": 731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YzN4Zj9S6so.mp4", "duration": 157.62413333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment that uses a cartoon to explain the concept of placing items within reach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment that uses a cartoon to explain the concept of placing items within reach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 114.0], "base_se_norm": [0.14195716381072998, 0.44647908210754395], "gt_se_raw": [89.0, 114.0], "gt_se_norm": [0.12926876544952393, 0.44647908210754395], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000731.pt"}
{"idx": 733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1t7VOZCyuJ0.mp4", "duration": 73.72, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man and woman are arguing with each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man and woman are arguing with each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 20.0], "base_se_norm": [-0.7015734910964966, -0.4574064016342163], "gt_se_raw": [11.0, 20.0], "gt_se_norm": [-0.7015734910964966, -0.4574064016342163], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000733.pt"}
{"idx": 735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1t7VOZCyuJ0.mp4", "duration": 73.72, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man stands up and picks up the stool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man stands up and picks up the stool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.15897995233535767, -0.07759088277816772], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [-0.15897995233535767, -0.07759088277816772], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000735.pt"}
{"idx": 737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1t7VOZCyuJ0.mp4", "duration": 73.72, "prompt_text": "Please find the visual event described by the sentence 'When does the slogan 'EMBRACE DISABLED TALENT, DON'T EXCLUDE IT.' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the slogan 'EMBRACE DISABLED TALENT, DON'T EXCLUDE IT.' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 48.0], "base_se_norm": [-0.05046123266220093, 0.30222463607788086], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.07759088277816772, 0.0037981271743774414], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000737.pt"}
{"idx": 739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WdC4DaYIeQ.mp4", "duration": 110.48537499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the snowman and the reindeer are fighting over the carrot with their mouths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the snowman and the reindeer are fighting over the carrot with their mouths.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 63.0], "base_se_norm": [-0.0768008828163147, 0.14042246341705322], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [-0.04059702157974243, -0.004393100738525391], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000739.pt"}
{"idx": 741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WdC4DaYIeQ.mp4", "duration": 110.48537499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the reindeer put the carrot nose back on the snowman's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the reindeer put the carrot nose back on the snowman's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 73.0], "base_se_norm": [0.28523802757263184, 0.3214418888092041], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [0.28523802757263184, 0.3214418888092041], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000741.pt"}
{"idx": 743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_RCAWNlJZCE.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the close-up of a man's hands pressing buttons on a calculator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the close-up of a man's hands pressing buttons on a calculator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.8989773988723755], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8989773988723755], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000743.pt"}
{"idx": 745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_RCAWNlJZCE.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man stands alone looking perplexed after his daughter walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man stands alone looking perplexed after his daughter walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 35.0], "base_se_norm": [-0.22549360990524292, 0.17859673500061035], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [-0.191819429397583, -0.057122647762298584], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000745.pt"}
{"idx": 747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_RCAWNlJZCE.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the end screen with the text 'SUBSCRIBE TO DAD'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the end screen with the text 'SUBSCRIBE TO DAD'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 59.0], "base_se_norm": [0.3806418180465698, 0.9867773056030273], "gt_se_raw": [42.0, 59.0], "gt_se_norm": [0.4143160581588745, 0.9867773056030273], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000747.pt"}
{"idx": 749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elXXLAiktZc.mp4", "duration": 87.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl turns the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl turns the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.3795480728149414, -0.26464956998825073], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.5174262523651123, -0.4714668393135071], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000749.pt"}
{"idx": 751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/elXXLAiktZc.mp4", "duration": 87.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the interval where the girl touches her doll's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the interval where the girl touches her doll's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.2186901569366455, -0.08081191778182983], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.2186901569366455, -0.12677133083343506], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000751.pt"}
{"idx": 753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9ZTX302mNw4.mp4", "duration": 19.6, "prompt_text": "Please find the visual event described by the sentence 'When does the boy's face first transform into an older man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy's face first transform into an older man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.6938775777816772, -0.4897959232330322], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.5918367505073547, -0.4897959232330322], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000753.pt"}
{"idx": 755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9ZTX302mNw4.mp4", "duration": 19.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the son sings the line 'Yes, dad'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the son sings the line 'Yes, dad'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [0.22448980808258057, 0.42857134342193604], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [0.3265305757522583, 0.42857134342193604], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000755.pt"}
{"idx": 757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9mmmipGP1h0.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'When are three men discussing and inspecting several large HVAC units in a workshop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three men discussing and inspecting several large HVAC units in a workshop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.6270396113395691], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.6270396113395691], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000757.pt"}
{"idx": 759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9mmmipGP1h0.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'At what time is a young man in safety glasses holding a device to measure the output of an air conditioner?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a young man in safety glasses holding a device to measure the output of an air conditioner?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.20079916715621948, 0.11888110637664795], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.20079916715621948, 0.11888110637664795], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000759.pt"}
{"idx": 761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9mmmipGP1h0.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a woman is shown working on an outdoor HVAC unit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a woman is shown working on an outdoor HVAC unit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [0.5451215505599976, 0.8648018836975098], "gt_se_raw": [29.0, 35.0], "gt_se_norm": [0.5451215505599976, 0.8648018836975098], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000761.pt"}
{"idx": 763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KYazqIHYqNI.mp4", "duration": 195.46193333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the woman performing CPR on the puppy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman performing CPR on the puppy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 56.0], "base_se_norm": [-0.5293201208114624, -0.4269983768463135], "gt_se_raw": [46.0, 51.0], "gt_se_norm": [-0.5293201208114624, -0.47815924882888794], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000763.pt"}
{"idx": 765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KYazqIHYqNI.mp4", "duration": 195.46193333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two puppies sleeping together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two puppies sleeping together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 99.0], "base_se_norm": [-0.027943670749664307, 0.012984991073608398], "gt_se_raw": [95.0, 99.0], "gt_se_norm": [-0.027943670749664307, 0.012984991073608398], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000765.pt"}
{"idx": 767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h8u_mj2IOcs.mp4", "duration": 54.854800000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the woman with blonde hair resting her chin on her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman with blonde hair resting her chin on her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 34.0], "base_se_norm": [-1.0, 0.23963630199432373], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9635401368141174, -0.7447807788848877], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000767.pt"}
{"idx": 769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h8u_mj2IOcs.mp4", "duration": 54.854800000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point is the page of the book being turned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the page of the book being turned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [0.2031763792037964, 0.31255602836608887], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [0.16671645641326904, 0.31255602836608887], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000769.pt"}
{"idx": 771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h8u_mj2IOcs.mp4", "duration": 54.854800000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the teacher uses her hand to cover her mouth while speaking to the student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the teacher uses her hand to cover her mouth while speaking to the student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 34.0], "base_se_norm": [-1.0, 0.23963630199432373], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.7083208560943604, -0.4895615577697754], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000771.pt"}
{"idx": 773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C97ZSYzwabo.mp4", "duration": 36.236000000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child pick up the orange toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child pick up the orange toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.5584501624107361, -0.33767521381378174], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.5032564401626587, -0.33767521381378174], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000773.pt"}
{"idx": 775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C97ZSYzwabo.mp4", "duration": 36.236000000000004, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the child covers their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the child covers their face with their hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [0.2142620086669922, 0.3246495723724365], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.2142620086669922, 0.3246495723724365], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000775.pt"}
{"idx": 777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmvB4oKrF1c.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'When are two men sitting and playing musical instruments in a recording booth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men sitting and playing musical instruments in a recording booth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.6270396113395691], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.6270396113395691], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000777.pt"}
{"idx": 779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmvB4oKrF1c.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'When is a band with a double bass performing on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a band with a double bass performing on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.20079916715621948, 0.11888110637664795], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.20079916715621948, 0.11888110637664795], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000779.pt"}
{"idx": 781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dmvB4oKrF1c.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Identify the close-up shot of a man playing an acoustic guitar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the close-up shot of a man playing an acoustic guitar.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [0.5451215505599976, 0.8648018836975098], "gt_se_raw": [29.0, 35.0], "gt_se_norm": [0.5451215505599976, 0.8648018836975098], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000781.pt"}
{"idx": 783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fUybrJmFv0c.mp4", "duration": 184.75123333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the anatomy of a butterfly is being explained with text labels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the anatomy of a butterfly is being explained with text labels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 50.0], "base_se_norm": [-0.6211121082305908, -0.45873165130615234], "gt_se_raw": [35.0, 49.0], "gt_se_norm": [-0.6211121082305908, -0.469556987285614], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000783.pt"}
{"idx": 785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fUybrJmFv0c.mp4", "duration": 184.75123333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a furry, orange and black caterpillar is crawling on a log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a furry, orange and black caterpillar is crawling on a log.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [-0.16644668579101562, -0.1231452226638794], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.16644668579101562, -0.1231452226638794], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000785.pt"}
{"idx": 787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XAyP11gQ22w.mp4", "duration": 171.43793333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the green hoodie demonstrate how she used the jammed stapler?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the green hoodie demonstrate how she used the jammed stapler?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 20.0], "base_se_norm": [-0.8366755843162537, -0.7666794061660767], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.8133435249328613, -0.790011465549469], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000787.pt"}
{"idx": 789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XAyP11gQ22w.mp4", "duration": 171.43793333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the green hoodie hold up the completed paper ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the green hoodie hold up the completed paper ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 45.0], "base_se_norm": [-0.6383531093597412, -0.4750286340713501], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.5333588123321533, -0.510026752948761], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000789.pt"}
{"idx": 791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XAyP11gQ22w.mp4", "duration": 171.43793333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where both women are smiling at each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where both women are smiling at each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.8250095844268799, -0.7666794061660767], "gt_se_raw": [95.0, 97.0], "gt_se_norm": [0.10827279090881348, 0.13160490989685059], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000791.pt"}
{"idx": 793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L0SPXFTfVRY.mp4", "duration": 36.302933333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the boy has his fingers in his mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the boy has his fingers in his mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, 0.9833108186721802], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [0.4874831438064575, 0.8731268644332886], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000793.pt"}
{"idx": 795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L0SPXFTfVRY.mp4", "duration": 36.302933333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman's head is close to the boy's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman's head is close to the boy's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 36.0], "base_se_norm": [-1.0, 0.9833108186721802], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.6143562197685242, -0.33889639377593994], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000795.pt"}
{"idx": 797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0g9_bCAb9AM.mp4", "duration": 154.38756666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe two women sitting at a table and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe two women sitting at a table and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 74.0], "base_se_norm": [-0.6761397123336792, -0.041373610496520996], "gt_se_raw": [25.0, 53.0], "gt_se_norm": [-0.6761397123336792, -0.3134162425994873], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000797.pt"}
{"idx": 799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0g9_bCAb9AM.mp4", "duration": 154.38756666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the train display board at the station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the train display board at the station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [-0.041373610496520996, 0.010444045066833496], "gt_se_raw": [74.0, 79.0], "gt_se_norm": [-0.041373610496520996, 0.023398399353027344], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000799.pt"}
{"idx": 801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0g9_bCAb9AM.mp4", "duration": 154.38756666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when we see a view of the street at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when we see a view of the street at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 130.0], "base_se_norm": [0.6322556734085083, 0.6840733289718628], "gt_se_raw": [126.0, 131.0], "gt_se_norm": [0.6322556734085083, 0.6970278024673462], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000801.pt"}
{"idx": 803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_MC3XuMvsDI.mp4", "duration": 130.38025, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the father's emotions are watching a soccer game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the father's emotions are watching a soccer game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.6318460702896118, -0.5398075580596924], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.6165063381195068, -0.5858268737792969], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000803.pt"}
{"idx": 805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_MC3XuMvsDI.mp4", "duration": 130.38025, "prompt_text": "Please find the visual event described by the sentence 'When does the red alarm light start flashing in the father's headquarters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the red alarm light start flashing in the father's headquarters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 84.0], "base_se_norm": [0.2578592300415039, 0.28853869438171387], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [0.2578592300415039, 0.28853869438171387], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000805.pt"}
{"idx": 807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/04pcSyzwoTg.mp4", "duration": 89.04, "prompt_text": "Please find the visual event described by the sentence 'When is the term 'the school to prison pipeline' first introduced and defined?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the term 'the school to prison pipeline' first introduced and defined?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.9550763964653015, -0.7529200315475464], "gt_se_raw": [3.0, 11.0], "gt_se_norm": [-0.9326145648956299, -0.7529200315475464], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000807.pt"}
{"idx": 809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/04pcSyzwoTg.mp4", "duration": 89.04, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment that states dropping out of high school makes a person three times more likely to be arrested.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment that states dropping out of high school makes a person three times more likely to be arrested.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.2362983226776123, -0.14645105600357056], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.2362983226776123, -0.1689128279685974], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000809.pt"}
{"idx": 811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/04pcSyzwoTg.mp4", "duration": 89.04, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a hand in the shape of a gun is pointed at a student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a hand in the shape of a gun is pointed at a student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8652291297912598, -0.7529200315475464], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [0.34770894050598145, 0.39263248443603516], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000811.pt"}
{"idx": 813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sdUUx5FdySs.mp4", "duration": 189.55603333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point is the kiwi hammering nails into the cliffside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the kiwi hammering nails into the cliffside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 74.0], "base_se_norm": [-0.40914565324783325, -0.21922820806503296], "gt_se_raw": [60.0, 72.0], "gt_se_norm": [-0.36694180965423584, -0.24033015966415405], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000813.pt"}
{"idx": 815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sdUUx5FdySs.mp4", "duration": 189.55603333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the kiwi puts on pilot goggles and prepares to jump.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the kiwi puts on pilot goggles and prepares to jump.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 130.0], "base_se_norm": [0.29776930809020996, 0.3716261386871338], "gt_se_raw": [98.0, 110.0], "gt_se_norm": [0.03399503231048584, 0.1606067419052124], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000815.pt"}
{"idx": 817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJcE1mqSJDA.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'When do the two men approach the wooden door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two men approach the wooden door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9000304937362671], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9000304937362671], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000817.pt"}
{"idx": 819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJcE1mqSJDA.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a pink shirt hand a card to a woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a pink shirt hand a card to a woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [0.1663111448287964, 0.2662806510925293], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.1663111448287964, 0.23295748233795166], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000819.pt"}
{"idx": 821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJcE1mqSJDA.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the website 'GET YOUR V CARD .ORG' is shown on a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the website 'GET YOUR V CARD .ORG' is shown on a tablet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [0.5328660011291504, 0.7328051328659058], "gt_se_raw": [46.0, 52.0], "gt_se_norm": [0.5328660011291504, 0.7328051328659058], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000821.pt"}
{"idx": 823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oOOCx287v4Q.mp4", "duration": 50.84245833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman and a boy are playing hand games on the lawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman and a boy are playing hand games on the lawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.6853023767471313, -0.40994197130203247], "gt_se_raw": [8.0, 15.0], "gt_se_norm": [-0.6853023767471313, -0.40994197130203247], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000823.pt"}
{"idx": 825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oOOCx287v4Q.mp4", "duration": 50.84245833333333, "prompt_text": "Please find the visual event described by the sentence 'Please show the part where the woman and the boy are communicating with sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please show the part where the woman and the boy are communicating with sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.05590713024139404, 0.2587904930114746], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.05590713024139404, 0.2587904930114746], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000825.pt"}
{"idx": 827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9uP7WjixVFM.mp4", "duration": 119.49437499999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man holding the bottle with both hands and showing the label?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man holding the bottle with both hands and showing the label?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 24.0], "base_se_norm": [-0.7656793594360352, -0.5983074903488159], "gt_se_raw": [15.0, 23.0], "gt_se_norm": [-0.7489421367645264, -0.6150446534156799], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000827.pt"}
{"idx": 829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9uP7WjixVFM.mp4", "duration": 119.49437499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the time range for the purple animated logo screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range for the purple animated logo screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8995768427848816, -0.815890908241272], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.8828396797180176, -0.815890908241272], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000829.pt"}
{"idx": 831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9uP7WjixVFM.mp4", "duration": 119.49437499999999, "prompt_text": "Please find the visual event described by the sentence 'When does the man drink the last few drops and finish the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man drink the last few drops and finish the bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 99.0], "base_se_norm": [0.5900329351425171, 0.6569817066192627], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [0.6235073804855347, 0.6737189292907715], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000831.pt"}
{"idx": 833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BXx51SR_3Sk.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the green jacket pick up the basketball that rolled towards him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the green jacket pick up the basketball that rolled towards him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [-0.6764551401138306, -0.6494930386543274], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.6629740595817566, -0.6360119581222534], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000833.pt"}
{"idx": 835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BXx51SR_3Sk.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man in the green jacket bounces the basketball off the other man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man in the green jacket bounces the basketball off the other man's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 65.0], "base_se_norm": [-0.15069466829299927, -0.1237325668334961], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [-0.15069466829299927, -0.1102515459060669], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000835.pt"}
{"idx": 837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IAq8pEFNeJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a man in a red sweatshirt is standing next to an excavator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a man in a red sweatshirt is standing next to an excavator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000837.pt"}
{"idx": 839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IAq8pEFNeJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When are three people seen working on a yellow dump truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three people seen working on a yellow dump truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08288431167602539, 0.3756735324859619], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000839.pt"}
{"idx": 841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IAq8pEFNeJs.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man with a beard and a red hoodie looks directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man with a beard and a red hoodie looks directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3756735324859619, 0.8342313766479492], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000841.pt"}
{"idx": 843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgQVj4iMm8Y.mp4", "duration": 184.88488488488488, "prompt_text": "Please find the visual event described by the sentence 'At what point does the blonde woman in a grey shirt laugh out loud?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the blonde woman in a grey shirt laugh out loud?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.7620140910148621, -0.729561448097229], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.7620140910148621, -0.729561448097229], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000843.pt"}
{"idx": 845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgQVj4iMm8Y.mp4", "duration": 184.88488488488488, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man in the brown leather jacket gives the middle finger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 171.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man in the brown leather jacket gives the middle finger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 171.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 171.0], "base_se_norm": [0.8281645774841309, 0.849799633026123], "gt_se_raw": [169.0, 171.0], "gt_se_norm": [0.8281645774841309, 0.849799633026123], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000845.pt"}
{"idx": 847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1awM6kmpd2g.mp4", "duration": 144.6445, "prompt_text": "Please find the visual event described by the sentence 'Show me when a group of friends is taking a photo inside the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a group of friends is taking a photo inside the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 25.0], "base_se_norm": [-0.8617299795150757, -0.6543248891830444], "gt_se_raw": [18.0, 25.0], "gt_se_norm": [-0.7511139512062073, -0.6543248891830444], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000847.pt"}
{"idx": 849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1awM6kmpd2g.mp4", "duration": 144.6445, "prompt_text": "Please find the visual event described by the sentence 'When can we see the digital rendering of the new multi-story car park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the digital rendering of the new multi-story car park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 72.0], "base_se_norm": [-0.11507177352905273, -0.004455745220184326], "gt_se_raw": [65.0, 72.0], "gt_se_norm": [-0.10124480724334717, -0.004455745220184326], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000849.pt"}
{"idx": 851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1awM6kmpd2g.mp4", "duration": 144.6445, "prompt_text": "Please find the visual event described by the sentence 'At what time is the orange 'Pendle Hill' sign shown on a brick wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the orange 'Pendle Hill' sign shown on a brick wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 113.0], "base_se_norm": [0.5347974300384521, 0.5624513626098633], "gt_se_raw": [110.0, 112.0], "gt_se_norm": [0.520970344543457, 0.5486243963241577], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000851.pt"}
{"idx": 853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jUPGpJ4NRkg.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman sells a hat from her street cart during winter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman sells a hat from her street cart during winter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.6336996555328369, -0.5004994869232178], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.6003996133804321, -0.5337995290756226], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000853.pt"}
{"idx": 855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jUPGpJ4NRkg.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'When does the robot's arm display 'Cash Flow Positive'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the robot's arm display 'Cash Flow Positive'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [0.2321012020111084, 0.33200132846832275], "gt_se_raw": [38.0, 39.0], "gt_se_norm": [0.2654012441635132, 0.29870128631591797], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000855.pt"}
{"idx": 857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9Byx6TxOPx4.mp4", "duration": 59.7597, "prompt_text": "Please find the visual event described by the sentence 'When can we see the student raising both hands to the sides of her head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the student raising both hands to the sides of her head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9330652356147766, -0.8326631784439087], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9665326476097107, -0.8995978832244873], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000857.pt"}
{"idx": 859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9Byx6TxOPx4.mp4", "duration": 59.7597, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher use the pointer to indicate something on the pink tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher use the pointer to indicate something on the pink tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [0.8741726875305176, 0.9745748043060303], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.09638100862503052, 0.004021048545837402], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000859.pt"}
{"idx": 861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9Byx6TxOPx4.mp4", "duration": 59.7597, "prompt_text": "Please find the visual event described by the sentence 'Find the instance where the teacher touches the student's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the instance where the teacher touches the student's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.8995978832244873, -0.7991957664489746], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [0.9076400995254517, 0.9745748043060303], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000861.pt"}
{"idx": 863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dXyrxjIcWdw.mp4", "duration": 77.49408333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the man look into the window on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man look into the window on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.4064062833786011, -0.2773641347885132], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.3289809823036194, -0.25155574083328247], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000863.pt"}
{"idx": 865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dXyrxjIcWdw.mp4", "duration": 77.49408333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man successfully opens the door and walks inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man successfully opens the door and walks inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.09670519828796387, 0.10976219177246094], "gt_se_raw": [36.0, 45.0], "gt_se_norm": [-0.07089674472808838, 0.16137909889221191], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000865.pt"}
{"idx": 867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ff5aCoq6xn8.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when a blue binder with the label 'COUNTING BOOK' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a blue binder with the label 'COUNTING BOOK' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.7355099320411682, -0.5702036619186401], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.7355099320411682, -0.5702036619186401], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000867.pt"}
{"idx": 869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ff5aCoq6xn8.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the '2 Straws' page is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the '2 Straws' page is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [0.09102153778076172, 0.256327748298645], "gt_se_raw": [28.0, 38.0], "gt_se_norm": [-0.07428479194641113, 0.256327748298645], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000869.pt"}
{"idx": 871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ff5aCoq6xn8.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher help the child count the toothbrushes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher help the child count the toothbrushes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 59.0], "base_se_norm": [0.6530628204345703, 0.9506142139434814], "gt_se_raw": [49.0, 60.0], "gt_se_norm": [0.6200016736984253, 0.983675479888916], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000871.pt"}
{"idx": 873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wDNxuom6Ge4.mp4", "duration": 68.4684, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the teacher is demonstrating 'lettuce' with a physical model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the teacher is demonstrating 'lettuce' with a physical model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 27.0], "base_se_norm": [-0.5326310992240906, -0.21131497621536255], "gt_se_raw": [17.0, 29.0], "gt_se_norm": [-0.5034205913543701, -0.15289390087127686], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000873.pt"}
{"idx": 875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wDNxuom6Ge4.mp4", "duration": 68.4684, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher interact with the student on the right using the yellow banana toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher interact with the student on the right using the yellow banana toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 67.0], "base_se_norm": [0.665001630783081, 0.9571073055267334], "gt_se_raw": [58.0, 66.0], "gt_se_norm": [0.6942121982574463, 0.9278967380523682], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000875.pt"}
{"idx": 877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1g4AUYiz0LU.mp4", "duration": 16.541666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the exact moment the child successfully claps their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the exact moment the child successfully claps their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.758186399936676, -0.516372799873352], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.8790931701660156, -0.758186399936676], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000877.pt"}
{"idx": 879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1g4AUYiz0LU.mp4", "duration": 16.541666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman singing 'If you're happy and you know it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman singing 'If you're happy and you know it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, 0.9345089197158813], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [0.4508817195892334, 0.6926952600479126], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000879.pt"}
{"idx": 881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1g4AUYiz0LU.mp4", "duration": 16.541666666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the period when the woman in the striped shirt is holding her hands together in a prayer-like gesture, waiting for the child's response.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the period when the woman in the striped shirt is holding her hands together in a prayer-like gesture, waiting for the child's response.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.2745591402053833, 0.4508817195892334], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.1536523699760437, 0.20906805992126465], "start_token_span": [83, 86], "end_token_span": [88, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000881.pt"}
{"idx": 883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nfosQcd4l5Q.mp4", "duration": 18.585233333333335, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a woman in a red T-shirt is visible in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a woman in a red T-shirt is visible in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 18.0], "base_se_norm": [-1.0, 0.9370216131210327], "gt_se_raw": [0.0, 18.0], "gt_se_norm": [-1.0, 0.9370216131210327], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000883.pt"}
{"idx": 885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nfosQcd4l5Q.mp4", "duration": 18.585233333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a yellow sweater touch the boy's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a yellow sweater touch the boy's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [0.07612311840057373, 0.5065723657608032], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [0.2913477420806885, 0.6141847372055054], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000885.pt"}
{"idx": 887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tloBMf_KmX4.mp4", "duration": 175.24, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the girl in the wheelchair going up the ramp to the station?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the girl in the wheelchair going up the ramp to the station?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [-0.0755535364151001, -0.041314780712127686], "gt_se_raw": [81.0, 85.0], "gt_se_norm": [-0.0755535364151001, -0.029901862144470215], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000887.pt"}
{"idx": 889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tloBMf_KmX4.mp4", "duration": 175.24, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man wearing headphones and riding the escalator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man wearing headphones and riding the escalator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 38.0], "base_se_norm": [-0.6347866058349609, -0.5663090944290161], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.6233736276626587, -0.5663090944290161], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000889.pt"}
{"idx": 891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tloBMf_KmX4.mp4", "duration": 175.24, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the man's prosthetic legs while he is sitting on the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the man's prosthetic legs while he is sitting on the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [-0.417941153049469, -0.39511531591415405], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [-0.42935407161712646, -0.39511531591415405], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000891.pt"}
{"idx": 893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LrI4FmRIHpI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When are two men seen inside the cab of a tractor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men seen inside the cab of a tractor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5409404635429382, -0.08188092708587646], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000893.pt"}
{"idx": 895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LrI4FmRIHpI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do two women walk down a set of outdoor stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two women walk down a set of outdoor stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000895.pt"}
{"idx": 897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zFCNUW0TfqE.mp4", "duration": 162.0, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the white coat appear to be in distress and holding her stomach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the white coat appear to be in distress and holding her stomach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 13.0], "base_se_norm": [-0.9753086566925049, -0.8395061492919922], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9629629850387573, -0.9135802388191223], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000897.pt"}
{"idx": 899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zFCNUW0TfqE.mp4", "duration": 162.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman drinks a glass of water while two other doctors look on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman drinks a glass of water while two other doctors look on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 76.0], "base_se_norm": [-0.18518519401550293, -0.06172841787338257], "gt_se_raw": [67.0, 72.0], "gt_se_norm": [-0.17283952236175537, -0.1111111044883728], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000899.pt"}
{"idx": 901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zFCNUW0TfqE.mp4", "duration": 162.0, "prompt_text": "Please find the visual event described by the sentence 'When does the woman vomit on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman vomit on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 47.0], "base_se_norm": [-0.4444444179534912, -0.4197530746459961], "gt_se_raw": [127.0, 131.0], "gt_se_norm": [0.5679012537002563, 0.6172839403152466], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000901.pt"}
{"idx": 903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/abPLDLV8O4s.mp4", "duration": 102.96, "prompt_text": "Please find the visual event described by the sentence 'When does the cat use a fireplace poker to knock over the Santa toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat use a fireplace poker to knock over the Santa toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.3589743375778198, -0.2812742590904236], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [-0.3589743375778198, -0.2812742590904236], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000903.pt"}
{"idx": 905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/abPLDLV8O4s.mp4", "duration": 102.96, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the cat is playing with the spilled food on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the cat is playing with the spilled food on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 78.0], "base_se_norm": [-0.1647241711616516, 0.5151515007019043], "gt_se_raw": [61.0, 78.0], "gt_se_norm": [0.18492615222930908, 0.5151515007019043], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000905.pt"}
{"idx": 907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MXVz4izTWZY.mp4", "duration": 145.47881214547883, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a person walking down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a person walking down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 20.0], "base_se_norm": [-0.9037660360336304, -0.7250458598136902], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.9037660360336304, -0.8350275158882141], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000907.pt"}
{"idx": 909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MXVz4izTWZY.mp4", "duration": 145.47881214547883, "prompt_text": "Please find the visual event described by the sentence 'At what point is a hospital bed visible inside an elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a hospital bed visible inside an elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 49.0], "base_se_norm": [-0.3813531994819641, -0.32636237144470215], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [-0.3813531994819641, -0.32636237144470215], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000909.pt"}
{"idx": 911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MXVz4izTWZY.mp4", "duration": 145.47881214547883, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a group of people get into a crowded elevator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a group of people get into a crowded elevator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 128.0], "base_se_norm": [0.42976152896881104, 0.7597064971923828], "gt_se_raw": [105.0, 128.0], "gt_se_norm": [0.44350922107696533, 0.7597064971923828], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000911.pt"}
{"idx": 913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W6gxb8IunvI.mp4", "duration": 59.62623333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the instructor's hand places the spiral notebook in the middle of the map.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the instructor's hand places the spiral notebook in the middle of the map.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.6310348510742188, -0.5304080247879028], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.6310348510742188, -0.5639503002166748], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000913.pt"}
{"idx": 915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W6gxb8IunvI.mp4", "duration": 59.62623333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is a blue label placed onto the spiral notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a blue label placed onto the spiral notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [0.4087759256362915, 0.47586047649383545], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [0.4087759256362915, 0.47586047649383545], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000915.pt"}
{"idx": 917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4O94zKHSgMU.mp4", "duration": 46.92, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman interacts with a 3D haptic sphere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman interacts with a 3D haptic sphere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 21.0], "base_se_norm": [-0.6589940190315247, -0.10485929250717163], "gt_se_raw": [9.0, 17.0], "gt_se_norm": [-0.616368293762207, -0.2753623127937317], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000917.pt"}
{"idx": 919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4O94zKHSgMU.mp4", "duration": 46.92, "prompt_text": "Please find the visual event described by the sentence 'When can we see a demonstration of a shape rotating in real-time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a demonstration of a shape rotating in real-time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [0.32139813899993896, 0.5345269441604614], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [0.36402392387390137, 0.5345269441604614], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000919.pt"}
{"idx": 921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4O94zKHSgMU.mp4", "duration": 46.92, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the animation that illustrates how focused ultrasound creates haptic feedback.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the animation that illustrates how focused ultrasound creates haptic feedback.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.8294969797134399, -0.6589940190315247], "gt_se_raw": [3.0, 9.0], "gt_se_norm": [-0.8721227645874023, -0.616368293762207], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000921.pt"}
{"idx": 923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mo-U0b64hA.mp4", "duration": 80.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of stirring the coffee grounds and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of stirring the coffee grounds and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.5249999761581421, -0.3999999761581421], "gt_se_raw": [18.0, 24.0], "gt_se_norm": [-0.550000011920929, -0.3999999761581421], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000923.pt"}
{"idx": 925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mo-U0b64hA.mp4", "duration": 80.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the plunger is being pushed down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the plunger is being pushed down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.10000002384185791, 0.0], "gt_se_raw": [36.0, 40.0], "gt_se_norm": [-0.10000002384185791, 0.0], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000925.pt"}
{"idx": 927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAeAqaA0Llg.mp4", "duration": 152.65249999999997, "prompt_text": "Please find the visual event described by the sentence 'When does the man write his name on the blackboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man write his name on the blackboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 45.0], "base_se_norm": [-0.5283404588699341, -0.41042560338974], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.515238881111145, -0.43662893772125244], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000927.pt"}
{"idx": 929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAeAqaA0Llg.mp4", "duration": 152.65249999999997, "prompt_text": "Please find the visual event described by the sentence 'At what point does the actual substitute teacher enter the classroom and start speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the actual substitute teacher enter the classroom and start speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 30.0], "base_se_norm": [-0.8034752011299133, -0.6069504022598267], "gt_se_raw": [94.0, 104.0], "gt_se_norm": [0.23155534267425537, 0.3625718355178833], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000929.pt"}
{"idx": 931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAeAqaA0Llg.mp4", "duration": 152.65249999999997, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a student in a green and white jacket pushes past another student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a student in a green and white jacket pushes past another student.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 11.0], "base_se_norm": [-0.934491753578186, -0.8558818101882935], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.934491753578186, -0.8951867818832397], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000931.pt"}
{"idx": 933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qOZNzVcq2IY.mp4", "duration": 174.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where icons for touch, hearing, and vision are displayed, with vision being crossed out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where icons for touch, hearing, and vision are displayed, with vision being crossed out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 31.0], "base_se_norm": [-0.7356828451156616, -0.6437464356422424], "gt_se_raw": [23.0, 31.0], "gt_se_norm": [-0.7356828451156616, -0.6437464356422424], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000933.pt"}
{"idx": 935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qOZNzVcq2IY.mp4", "duration": 174.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment explaining tactile ground surface indicators on the pavement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment explaining tactile ground surface indicators on the pavement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 90.0], "base_se_norm": [-0.1151120662689209, 0.03428459167480469], "gt_se_raw": [77.0, 90.0], "gt_se_norm": [-0.1151120662689209, 0.03428459167480469], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000935.pt"}
{"idx": 937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ke6s5UH9n2w.mp4", "duration": 162.42893333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player knock another player into the void?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player knock another player into the void?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 94.0], "base_se_norm": [0.12048995494842529, 0.15742921829223633], "gt_se_raw": [91.0, 93.0], "gt_se_norm": [0.12048995494842529, 0.14511609077453613], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000937.pt"}
{"idx": 939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ke6s5UH9n2w.mp4", "duration": 162.42893333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the hammer and sickle emblem visible on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the hammer and sickle emblem visible on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 50.0], "base_se_norm": [-0.4705377221107483, -0.3843461871147156], "gt_se_raw": [44.0, 92.0], "gt_se_norm": [-0.4582245945930481, 0.1328030824661255], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000939.pt"}
{"idx": 941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ke6s5UH9n2w.mp4", "duration": 162.42893333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the game ends and the 'VICTORY' message appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the game ends and the 'VICTORY' message appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 103.0], "base_se_norm": [0.206681489944458, 0.2682468891143799], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [0.206681489944458, 0.24362075328826904], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000941.pt"}
{"idx": 943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOKLZsyN-aw.mp4", "duration": 14.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog turn its head to face more towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog turn its head to face more towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.45205479860305786, -0.04109591245651245], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.45205479860305786, -0.31506848335266113], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000943.pt"}
{"idx": 945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOKLZsyN-aw.mp4", "duration": 14.6, "prompt_text": "Please find the visual event described by the sentence 'During which period is the dog wearing a green hat with a feather?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which period is the dog wearing a green hat with a feather?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.9178081750869751], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, 0.9178081750869751], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000945.pt"}
{"idx": 947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPQwSEzXBXU.mp4", "duration": 104.4, "prompt_text": "Please find the visual event described by the sentence 'When is the Braillo 300 embosser shown printing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the Braillo 300 embosser shown printing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 97.0], "base_se_norm": [0.8007662296295166, 0.8582375049591064], "gt_se_raw": [95.0, 98.0], "gt_se_norm": [0.8199234008789062, 0.8773945569992065], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000947.pt"}
{"idx": 949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPQwSEzXBXU.mp4", "duration": 104.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman explain what is on the shelves behind her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman explain what is on the shelves behind her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 53.0], "base_se_norm": [-0.48275864124298096, 0.015325665473937988], "gt_se_raw": [28.0, 35.0], "gt_se_norm": [-0.4636015295982361, -0.3295019268989563], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000949.pt"}
{"idx": 951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPQwSEzXBXU.mp4", "duration": 104.4, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of a braille embosser printing a page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of a braille embosser printing a page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.7701149582862854, -0.6934865713119507], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.7701149582862854, -0.6743295192718506], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000951.pt"}
{"idx": 953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0klQnTscgRs.mp4", "duration": 117.05026666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when two young women are smelling the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when two young women are smelling the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 24.0], "base_se_norm": [-0.6924397945404053, -0.5899198055267334], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.6924397945404053, -0.6240931749343872], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000953.pt"}
{"idx": 955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0klQnTscgRs.mp4", "duration": 117.05026666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where the woman is demonstrating how to replace the scent tank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where the woman is demonstrating how to replace the scent tank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 49.0], "base_se_norm": [-0.24818629026412964, -0.16275286674499512], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.24818629026412964, -0.17983955144882202], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000955.pt"}
{"idx": 957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Zh2-rVsXWUU.mp4", "duration": 107.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the superheroes rescuing people from the burning building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the superheroes rescuing people from the burning building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 43.0], "base_se_norm": [-0.9253174066543579, -0.19716209173202515], "gt_se_raw": [4.0, 43.0], "gt_se_norm": [-0.9253174066543579, -0.19716209173202515], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000957.pt"}
{"idx": 959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Zh2-rVsXWUU.mp4", "duration": 107.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the men in ski masks are standing over unconscious people in a jewelry store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the men in ski masks are standing over unconscious people in a jewelry store.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 62.0], "base_se_norm": [-0.12247949838638306, 0.15758025646209717], "gt_se_raw": [60.0, 69.0], "gt_se_norm": [0.12023890018463135, 0.28827476501464844], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000959.pt"}
{"idx": 961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Zh2-rVsXWUU.mp4", "duration": 107.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point do the police officers discover their colleague has been frozen solid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the police officers discover their colleague has been frozen solid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 106.0], "base_se_norm": [0.8483942747116089, 0.9790887832641602], "gt_se_raw": [98.0, 107.0], "gt_se_norm": [0.829723596572876, 0.9977594614028931], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000961.pt"}
{"idx": 963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cT_Wuzag6VU.mp4", "duration": 129.66286666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is tying rocks to his hands with a tie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is tying rocks to his hands with a tie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 63.0], "base_se_norm": [-0.39844000339508057, -0.02824920415878296], "gt_se_raw": [40.0, 62.0], "gt_se_norm": [-0.38301539421081543, -0.043673813343048096], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000963.pt"}
{"idx": 965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cT_Wuzag6VU.mp4", "duration": 129.66286666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man is skateboarding past a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man is skateboarding past a waterfall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [0.5578862428665161, 0.6195846796035767], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [0.5578862428665161, 0.6195846796035767], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000965.pt"}
{"idx": 967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AMHWyS5GyeY.mp4", "duration": 101.68, "prompt_text": "Please find the visual event described by the sentence 'When does the man put the pet carrier on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man put the pet carrier on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 15.0], "base_se_norm": [-0.8623131513595581, -0.7049567699432373], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8623131513595581, -0.8033044934272766], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000967.pt"}
{"idx": 969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AMHWyS5GyeY.mp4", "duration": 101.68, "prompt_text": "Please find the visual event described by the sentence 'At what point does the small kitten come out of the carrier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the small kitten come out of the carrier?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 44.0], "base_se_norm": [-0.21321791410446167, -0.13453972339630127], "gt_se_raw": [46.0, 49.0], "gt_se_norm": [-0.09520065784454346, -0.03619199991226196], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000969.pt"}
{"idx": 971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AMHWyS5GyeY.mp4", "duration": 101.68, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment Simon's cat faints from shock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment Simon's cat faints from shock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 75.0], "base_se_norm": [0.3571990728378296, 0.4752163887023926], "gt_se_raw": [71.0, 74.0], "gt_se_norm": [0.3965381383895874, 0.45554685592651367], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000971.pt"}
{"idx": 973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzvK1lJwVUM.mp4", "duration": 133.2164111111111, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of sifting flour into the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of sifting flour into the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.6546971797943115, -0.5796313881874084], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.6697103977203369, -0.5946445465087891], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000973.pt"}
{"idx": 975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzvK1lJwVUM.mp4", "duration": 133.2164111111111, "prompt_text": "Please find the visual event described by the sentence 'At what point are the different types of cookie dough combined in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the different types of cookie dough combined in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 78.0], "base_se_norm": [0.02089524269104004, 0.1710268259048462], "gt_se_raw": [70.0, 78.0], "gt_se_norm": [0.05092155933380127, 0.1710268259048462], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000975.pt"}
{"idx": 977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ftde81QNXDY.mp4", "duration": 6.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Lisa look frustrated while talking to the doll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Lisa look frustrated while talking to the doll?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.4088670015335083], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.4088670015335083], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000977.pt"}
{"idx": 979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ftde81QNXDY.mp4", "duration": 6.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where a piano is visible in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where a piano is visible in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.11330050230026245, 0.7733989953994751], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.4088670015335083, 0.7733989953994751], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000979.pt"}
{"idx": 981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ftde81QNXDY.mp4", "duration": 6.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the doll is shown with a smiling expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the doll is shown with a smiling expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.11330050230026245, 0.7733989953994751], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.4088670015335083, 0.7733989953994751], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000981.pt"}
{"idx": 983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WwPXYlVwnk.mp4", "duration": 76.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the character in orange armor fire blue energy beams from her arms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the character in orange armor fire blue energy beams from her arms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.5, -0.44736844301223755], "gt_se_raw": [19.0, 20.0], "gt_se_norm": [-0.5, -0.4736841917037964], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000983.pt"}
{"idx": 985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-WwPXYlVwnk.mp4", "duration": 76.0, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the character's upgrades, like 'Buster Cannon +' and 'Thruster Boots', are displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the character's upgrades, like 'Buster Cannon +' and 'Thruster Boots', are displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 60.0], "base_se_norm": [0.4736841917037964, 0.5789474248886108], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [0.4736841917037964, 0.5526316165924072], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000985.pt"}
{"idx": 987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJry7As0rJU.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man wearing a black shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man wearing a black shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 36.0], "base_se_norm": [-0.6637977361679077, -0.5677399635314941], "gt_se_raw": [29.0, 36.0], "gt_se_norm": [-0.6517904996871948, -0.5677399635314941], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000987.pt"}
{"idx": 989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJry7As0rJU.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the text 'BIG SANDY COMMUNITY & TECHNICAL COLLEGE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the text 'BIG SANDY COMMUNITY & TECHNICAL COLLEGE'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.5557327270507812, -0.48368942737579346], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.5557327270507812, -0.48368942737579346], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000989.pt"}
{"idx": 991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJry7As0rJU.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man wearing a blue polo shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man wearing a blue polo shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 133.0], "base_se_norm": [0.5369246006011963, 0.5969606637954712], "gt_se_raw": [90.0, 96.0], "gt_se_norm": [0.08065009117126465, 0.15269339084625244], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000991.pt"}
{"idx": 993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUusX1Js6R0.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment Ernie demonstrates his special 'fish call' for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment Ernie demonstrates his special 'fish call' for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 61.0], "base_se_norm": [-0.5677399635314941, -0.2675594091415405], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [-0.20752328634262085, -0.15949440002441406], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000993.pt"}
{"idx": 995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUusX1Js6R0.mp4", "duration": 166.56640000000002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when Bert tries to call the fish but Ernie tells him it's too quiet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when Bert tries to call the fish but Ernie tells him it's too quiet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 42.0], "base_se_norm": [-0.5677399635314941, -0.4956966042518616], "gt_se_raw": [107.0, 114.0], "gt_se_norm": [0.2847728729248047, 0.36882340908050537], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000995.pt"}
{"idx": 997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0LHxvxdRnYc.mp4", "duration": 138.30483333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the green suit fight the old lady on the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the green suit fight the old lady on the train?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8409310579299927, -0.7686270475387573], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.8264702558517456, -0.7830878496170044], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000997.pt"}
{"idx": 999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0LHxvxdRnYc.mp4", "duration": 138.30483333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman is in the cockpit of a fighter jet, giving a shaka sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman is in the cockpit of a fighter jet, giving a shaka sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [-0.17573386430740356, -0.13235139846801758], "gt_se_raw": [59.0, 60.0], "gt_se_norm": [-0.14681220054626465, -0.13235139846801758], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00000999.pt"}
{"idx": 1001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0LHxvxdRnYc.mp4", "duration": 138.30483333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man pet the ginger cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man pet the ginger cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 131.0], "base_se_norm": [0.764218807220459, 0.8943661451339722], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.764218807220459, 0.8076012134552002], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001001.pt"}
{"idx": 1003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nr1IhBX8m1w.mp4", "duration": 28.5, "prompt_text": "Please find the visual event described by the sentence 'Show me when the train's headlight is visible in the distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the train's headlight is visible in the distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.3684210777282715, -0.08771932125091553], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.5087718963623047, -0.08771932125091553], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001003.pt"}
{"idx": 1005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nr1IhBX8m1w.mp4", "duration": 28.5, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the approaching train sounds its horn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the approaching train sounds its horn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.2982456088066101, -0.08771932125091553], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.2280701994895935, 0.12280702590942383], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001005.pt"}
{"idx": 1007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_B2UZ_ZoxU.mp4", "duration": 185.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl scribbling angrily in her diary?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl scribbling angrily in her diary?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 33.0], "base_se_norm": [-0.720367431640625, -0.6450817584991455], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.688102126121521, -0.6450817584991455], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001007.pt"}
{"idx": 1009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_B2UZ_ZoxU.mp4", "duration": 185.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the older sister gets off the school bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the older sister gets off the school bus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 64.0], "base_se_norm": [-0.3654492497444153, -0.31167376041412354], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [-0.3654492497444153, -0.34393906593322754], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001009.pt"}
{"idx": 1011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_B2UZ_ZoxU.mp4", "duration": 185.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When do the two sisters finally hug each other at the doorway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two sisters finally hug each other at the doorway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 173.0], "base_se_norm": [0.7638360261917114, 0.8606319427490234], "gt_se_raw": [164.0, 172.0], "gt_se_norm": [0.7638360261917114, 0.8498767614364624], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001011.pt"}
{"idx": 1013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J95dkdlWQbw.mp4", "duration": 94.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where everyone sits down on the sofa to watch TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where everyone sits down on the sofa to watch TV.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.6596951484680176, -0.574618935585022], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.6384260654449463, -0.574618935585022], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001013.pt"}
{"idx": 1015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J95dkdlWQbw.mp4", "duration": 94.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip on the TV where people are dancing at a party with balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip on the TV where people are dancing at a party with balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [-0.06416165828704834, 0.08472168445587158], "gt_se_raw": [44.0, 51.0], "gt_se_norm": [-0.06416165828704834, 0.08472168445587158], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001015.pt"}
{"idx": 1017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hywRdDVR76A.mp4", "duration": 135.52, "prompt_text": "Please find the visual event described by the sentence 'Show the segment that explains the chromosomal difference between males and females.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment that explains the chromosomal difference between males and females.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 63.0], "base_se_norm": [-0.18831169605255127, -0.07024794816970825], "gt_se_raw": [52.0, 63.0], "gt_se_norm": [-0.232585608959198, -0.07024794816970825], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001017.pt"}
{"idx": 1019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hywRdDVR76A.mp4", "duration": 135.52, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the basic parts of a cell are illustrated and labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the basic parts of a cell are illustrated and labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 36.0], "base_se_norm": [-0.6015348434448242, -0.46871310472488403], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.6015348434448242, -0.46871310472488403], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001019.pt"}
{"idx": 1021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hywRdDVR76A.mp4", "duration": 135.52, "prompt_text": "Please find the visual event described by the sentence 'When are different types of cells, such as nerve cells and skin cells, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are different types of cells, such as nerve cells and skin cells, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.7343565225601196, -0.6162928342819214], "gt_se_raw": [17.0, 24.0], "gt_se_norm": [-0.7491145133972168, -0.645808756351471], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001021.pt"}
{"idx": 1023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DIBw9dSVKdU.mp4", "duration": 126.25112499999999, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Ralph and Vanellope are launched into the internet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Ralph and Vanellope are launched into the internet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 46.0], "base_se_norm": [-0.5089152455329895, -0.271293580532074], "gt_se_raw": [32.0, 40.0], "gt_se_norm": [-0.49307382106781006, -0.3663422465324402], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001023.pt"}
{"idx": 1025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DIBw9dSVKdU.mp4", "duration": 126.25112499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the sorrowful kitten painting is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the sorrowful kitten painting is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 75.0], "base_se_norm": [0.07721817493438721, 0.18810820579528809], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [0.07721817493438721, 0.12474250793457031], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001025.pt"}
{"idx": 1027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DqilSuFK3B8.mp4", "duration": 438.6, "prompt_text": "Please find the visual event described by the sentence 'When are two women performing a choreographed dance in a dance studio?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women performing a choreographed dance in a dance studio?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, -0.9270405769348145], "gt_se_raw": [117.0, 134.0], "gt_se_norm": [-0.4664842486381531, -0.3889648914337158], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001027.pt"}
{"idx": 1029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DqilSuFK3B8.mp4", "duration": 438.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does a white dog appear on the woman's lap while she is using a laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a white dog appear on the woman's lap while she is using a laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 83.0], "base_se_norm": [-0.7081623077392578, -0.6215230226516724], "gt_se_raw": [64.0, 83.0], "gt_se_norm": [-0.7081623077392578, -0.6215230226516724], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001029.pt"}
{"idx": 1031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DqilSuFK3B8.mp4", "duration": 438.6, "prompt_text": "Please find the visual event described by the sentence 'Show the demonstration of how music is edited on the timeline in the editing software.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 242.0 - 257.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the demonstration of how music is edited on the timeline in the editing software.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 242.0 - 257.0 seconds.<|im_end|>\n", "base_se_raw": [242.0, 257.0], "base_se_norm": [0.10351121425628662, 0.17191064357757568], "gt_se_raw": [242.0, 257.0], "gt_se_norm": [0.10351121425628662, 0.17191064357757568], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001031.pt"}
{"idx": 1033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5L4DQfVIcdg.mp4", "duration": 314.875, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where the boy is visibly upset after the rocket ride malfunctions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where the boy is visibly upset after the rocket ride malfunctions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 133.0], "base_se_norm": [-0.4346963167190552, -0.15522032976150513], "gt_se_raw": [88.0, 100.0], "gt_se_norm": [-0.4410480260848999, -0.36482733488082886], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001033.pt"}
{"idx": 1035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5L4DQfVIcdg.mp4", "duration": 314.875, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy at the lemonade stand age from a child to an old man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy at the lemonade stand age from a child to an old man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 182.0], "base_se_norm": [0.060738444328308105, 0.1560143232345581], "gt_se_raw": [165.0, 172.0], "gt_se_norm": [0.0480349063873291, 0.09249699115753174], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001035.pt"}
{"idx": 1037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T52u9jDtIM4.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the university's main entrance sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the university's main entrance sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9730378985404968, -0.9325948357582092], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9730378985404968, -0.9460758566856384], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001037.pt"}
{"idx": 1039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T52u9jDtIM4.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the male student in the green polo shirt give his interview?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the male student in the green polo shirt give his interview?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 55.0], "base_se_norm": [-0.3259481191635132, -0.2585429549217224], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [-0.33942919969558716, -0.2585429549217224], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001039.pt"}
{"idx": 1041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T52u9jDtIM4.mp4", "duration": 148.35654166666666, "prompt_text": "Please find the visual event described by the sentence 'When are the 'YOU HAVE ARRIVED' pamphlets shown on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the 'YOU HAVE ARRIVED' pamphlets shown on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 92.0], "base_se_norm": [0.21329331398010254, 0.2402554750442505], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [0.19981229305267334, 0.2402554750442505], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001041.pt"}
{"idx": 1043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2hm0_sNiN2M.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'Show the part where someone is drying their hands with a paper towel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where someone is drying their hands with a paper towel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [0.43333327770233154, 0.5666667222976685], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [0.43333327770233154, 0.5666667222976685], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001043.pt"}
{"idx": 1045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2hm0_sNiN2M.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'When are the fingertips being scrubbed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the fingertips being scrubbed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 31.0], "base_se_norm": [-0.13333332538604736, 0.03333330154418945], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.13333332538604736, 0.03333330154418945], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001045.pt"}
{"idx": 1047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I8ZvdblLcnk.mp4", "duration": 63.998999999999995, "prompt_text": "Please find the visual event described by the sentence 'When is the woman helping the girl with the sock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman helping the girl with the sock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 63.0], "base_se_norm": [-1.0, 0.9687807559967041], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.8437475562095642], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001047.pt"}
{"idx": 1049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I8ZvdblLcnk.mp4", "duration": 63.998999999999995, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl laugh while trying to put on her sock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl laugh while trying to put on her sock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 35.0], "base_se_norm": [-0.15623682737350464, 0.09376704692840576], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [1.5616416931152344e-05, 0.09376704692840576], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001049.pt"}
{"idx": 1051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I8ZvdblLcnk.mp4", "duration": 63.998999999999995, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman gives the girl a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman gives the girl a hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.8437788486480713, 0.9687807559967041], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.8437788486480713, 0.9375302791595459], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001051.pt"}
{"idx": 1053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WoKaJfUHJmg.mp4", "duration": 166.09926666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where different accessibility services are displayed as a series of images.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where different accessibility services are displayed as a series of images.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 73.0], "base_se_norm": [-0.36182743310928345, -0.12100762128829956], "gt_se_raw": [54.0, 73.0], "gt_se_norm": [-0.3497864603996277, -0.12100762128829956], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001053.pt"}
{"idx": 1055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WoKaJfUHJmg.mp4", "duration": 166.09926666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part with the black and white photos of the Opera House during its construction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part with the black and white photos of the Opera House during its construction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 119.0], "base_se_norm": [0.3485910892486572, 0.4328780174255371], "gt_se_raw": [112.0, 119.0], "gt_se_norm": [0.3485910892486572, 0.4328780174255371], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001055.pt"}
{"idx": 1057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1lETr5TeT7E.mp4", "duration": 106.53976666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the little girl crouch down next to the fountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little girl crouch down next to the fountain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 33.0], "base_se_norm": [-0.6620979905128479, -0.3805130124092102], "gt_se_raw": [18.0, 35.0], "gt_se_norm": [-0.6620979905128479, -0.3429683446884155], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001057.pt"}
{"idx": 1059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1lETr5TeT7E.mp4", "duration": 106.53976666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl tap the pebbles inside the fountain with her cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl tap the pebbles inside the fountain with her cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [0.013705968856811523, 0.16388463973999023], "gt_se_raw": [53.0, 61.0], "gt_se_norm": [-0.005066335201263428, 0.14511239528656006], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001059.pt"}
{"idx": 1061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1lETr5TeT7E.mp4", "duration": 106.53976666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman takes the girl's hand and they walk away from the fountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman takes the girl's hand and they walk away from the fountain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 106.0], "base_se_norm": [0.6519653797149658, 0.9898673295974731], "gt_se_raw": [89.0, 102.0], "gt_se_norm": [0.6707377433776855, 0.9147779941558838], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001061.pt"}
{"idx": 1063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TU6VrllUQjw.mp4", "duration": 29.462766666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a man in a priest's collar is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a man in a priest's collar is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 27.0], "base_se_norm": [-0.25329482555389404, 0.8328217267990112], "gt_se_raw": [10.0, 27.0], "gt_se_norm": [-0.3211771249771118, 0.8328217267990112], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001063.pt"}
{"idx": 1065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TU6VrllUQjw.mp4", "duration": 29.462766666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time does the priest gesture with his hand while talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the priest gesture with his hand while talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [0.15399885177612305, 0.2897634506225586], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [0.15399885177612305, 0.35764575004577637], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001065.pt"}
{"idx": 1067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/olit-B5Yldc.mp4", "duration": 108.17391304347827, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a man screaming in fear of his backstory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a man screaming in fear of his backstory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.7781350612640381, -0.7411575317382812], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.7781350612640381, -0.7596462965011597], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001067.pt"}
{"idx": 1069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/olit-B5Yldc.mp4", "duration": 108.17391304347827, "prompt_text": "Please find the visual event described by the sentence 'At what point does a cow appear and invite others to go skateboarding?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a cow appear and invite others to go skateboarding?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 63.0], "base_se_norm": [-0.278938889503479, 0.16479098796844482], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.278938889503479, -0.22347265481948853], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001069.pt"}
{"idx": 1071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/olit-B5Yldc.mp4", "duration": 108.17391304347827, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a cow rides a skateboard through a scene of dead bodies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a cow rides a skateboard through a scene of dead bodies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 98.0], "base_se_norm": [0.737942099571228, 0.8118971586227417], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.737942099571228, 0.7934083938598633], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001071.pt"}
{"idx": 1073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZbCE6wzNbE.mp4", "duration": 105.8, "prompt_text": "Please find the visual event described by the sentence 'Show me when the screen reader audibly misinterprets the mathematical equation as an image file name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the screen reader audibly misinterprets the mathematical equation as an image file name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 39.0], "base_se_norm": [-0.4328922629356384, -0.2627599239349365], "gt_se_raw": [30.0, 40.0], "gt_se_norm": [-0.4328922629356384, -0.24385637044906616], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001073.pt"}
{"idx": 1075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZbCE6wzNbE.mp4", "duration": 105.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the diagram explaining the input and output process of MathML Cloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the diagram explaining the input and output process of MathML Cloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 61.0], "base_se_norm": [-0.03591686487197876, 0.15311908721923828], "gt_se_raw": [51.0, 61.0], "gt_se_norm": [-0.03591686487197876, 0.15311908721923828], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001075.pt"}
{"idx": 1077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G1LP4CwL0KE.mp4", "duration": 46.6, "prompt_text": "Please find the visual event described by the sentence 'Show me when the boy is adding black counters to the second ten-frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the boy is adding black counters to the second ten-frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.6566523313522339, -0.3562231659889221], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.6566523313522339, -0.3991416096687317], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001077.pt"}
{"idx": 1079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G1LP4CwL0KE.mp4", "duration": 46.6, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the number '19' is correctly assembled on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the number '19' is correctly assembled on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [0.6309013366699219, 0.8454936742782593], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [0.716738224029541, 0.8454936742782593], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001079.pt"}
{"idx": 1081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G1LP4CwL0KE.mp4", "duration": 46.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the boy picks the card with the number '9' and places it on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the boy picks the card with the number '9' and places it on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [0.6738197803497314, 0.8025751113891602], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [0.6738197803497314, 0.8025751113891602], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001081.pt"}
{"idx": 1083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TubxNbCQ4Fk.mp4", "duration": 171.90506666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a man is walking on the beach with a surfboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a man is walking on the beach with a surfboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.5113582015037537, -0.46482086181640625], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [-0.5113582015037537, -0.46482086181640625], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001083.pt"}
{"idx": 1085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TubxNbCQ4Fk.mp4", "duration": 171.90506666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man riding a bicycle on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man riding a bicycle on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 98.0], "base_se_norm": [0.11689555644989014, 0.14016425609588623], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.11689555644989014, 0.14016425609588623], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001085.pt"}
{"idx": 1087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aSq1cez_flQ.mp4", "duration": 133.71810027119116, "prompt_text": "Please find the visual event described by the sentence 'When are they playing the musical intro before the singing starts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are they playing the musical intro before the singing starts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, -0.7756474018096924], "gt_se_raw": [15.0, 23.0], "gt_se_norm": [-0.7756474018096924, -0.6559927463531494], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001087.pt"}
{"idx": 1089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aSq1cez_flQ.mp4", "duration": 133.71810027119116, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man sing his solo part?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man sing his solo part?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 63.0], "base_se_norm": [-0.2521580457687378, -0.05771917104721069], "gt_se_raw": [50.0, 68.0], "gt_se_norm": [-0.2521580457687378, 0.017065048217773438], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001089.pt"}
{"idx": 1091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h6oooSMEeQw.mp4", "duration": 106.87343333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the pyramid being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the pyramid being shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 57.0], "base_se_norm": [-0.15788239240646362, 0.06668221950531006], "gt_se_raw": [24.0, 45.0], "gt_se_norm": [-0.5508706569671631, -0.15788239240646362], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001091.pt"}
{"idx": 1093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h6oooSMEeQw.mp4", "duration": 106.87343333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 24.0], "base_se_norm": [-1.0, -0.5508706569671631], "gt_se_raw": [0.0, 24.0], "gt_se_norm": [-1.0, -0.5508706569671631], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001093.pt"}
{"idx": 1095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Co4dLH29PvM.mp4", "duration": 89.58949999999999, "prompt_text": "Please find the visual event described by the sentence 'When are two men walking along an outdoor covered walkway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men walking along an outdoor covered walkway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 24.0], "base_se_norm": [-0.5535191297531128, -0.46422290802001953], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.5535191297531128, -0.48654699325561523], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001095.pt"}
{"idx": 1097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Co4dLH29PvM.mp4", "duration": 89.58949999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a vest addressing someone sitting on the judge's bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a vest addressing someone sitting on the judge's bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [0.40641486644744873, 0.47338688373565674], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [0.384090781211853, 0.4510629177093506], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001097.pt"}
{"idx": 1099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Co4dLH29PvM.mp4", "duration": 89.58949999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man and a woman walking and talking in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man and a woman walking and talking in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.15168631076812744, -0.062390148639678955], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.1293622851371765, -0.062390148639678955], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001099.pt"}
{"idx": 1101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6wFr-o_M_Uo.mp4", "duration": 122.8, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the speaker mentions being legally blind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the speaker mentions being legally blind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.5602605938911438, -0.5114006996154785], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.7882736325263977, -0.7394136786460876], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001101.pt"}
{"idx": 1103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6wFr-o_M_Uo.mp4", "duration": 122.8, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the woman explains what NVDA stands for.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the woman explains what NVDA stands for.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.6742671132087708, -0.5114006996154785], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [-0.1856677532196045, -0.15309447050094604], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001103.pt"}
{"idx": 1105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DvaPRlZtfyc.mp4", "duration": 106.9068, "prompt_text": "Please find the visual event described by the sentence 'When does Mario fall out of his wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Mario fall out of his wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.8316290378570557, -0.7942132949829102], "gt_se_raw": [65.0, 66.0], "gt_se_norm": [0.21601247787475586, 0.2347203493118286], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001105.pt"}
{"idx": 1107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DvaPRlZtfyc.mp4", "duration": 106.9068, "prompt_text": "Please find the visual event described by the sentence 'At what point does Mario get a mushroom power-up while in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Mario get a mushroom power-up while in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 45.0], "base_se_norm": [-0.1955609917640686, -0.1581452488899231], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [-0.21426886320114136, -0.1581452488899231], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001107.pt"}
{"idx": 1109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DvaPRlZtfyc.mp4", "duration": 106.9068, "prompt_text": "Please find the visual event described by the sentence 'When can we see Luigi appear next to Mario?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Luigi appear next to Mario?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.5323028564453125, -0.4200555682182312], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.5135949850082397, -0.47617924213409424], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001109.pt"}
{"idx": 1111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKTQW5i9_p8.mp4", "duration": 125.44, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when she says 'act as if the house was on fire'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when she says 'act as if the house was on fire'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.7289540767669678, -0.6651785373687744], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.7289540767669678, -0.6811224222183228], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001111.pt"}
{"idx": 1113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zKTQW5i9_p8.mp4", "duration": 125.44, "prompt_text": "Please find the visual event described by the sentence 'At what point does Donald Trump walk past her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Donald Trump walk past her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 87.0], "base_se_norm": [0.3073979616165161, 0.38711726665496826], "gt_se_raw": [82.0, 87.0], "gt_se_norm": [0.3073979616165161, 0.38711726665496826], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001113.pt"}
{"idx": 1115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X-AldaPHQdE.mp4", "duration": 107.07363333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy peeling the egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy peeling the egg?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 48.0], "base_se_norm": [-0.4396379590034485, -0.10342073440551758], "gt_se_raw": [31.0, 47.0], "gt_se_norm": [-0.4209592342376709, -0.12209945917129517], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001115.pt"}
{"idx": 1117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X-AldaPHQdE.mp4", "duration": 107.07363333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the lighter used to set the paper on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the lighter used to set the paper on fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.10204541683197021, 0.17676031589508057], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [0.10204541683197021, 0.1394028663635254], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001117.pt"}
{"idx": 1119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X-AldaPHQdE.mp4", "duration": 107.07363333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the boy explains the science behind the experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the boy explains the science behind the experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 53.0], "base_se_norm": [-0.5517103672027588, -0.010027050971984863], "gt_se_raw": [75.0, 91.0], "gt_se_norm": [0.40090513229370117, 0.6997648477554321], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001119.pt"}
{"idx": 1121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRjCaSrmm2A.mp4", "duration": 53.13641666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man wearing a red plaid shirt gets excited about the game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man wearing a red plaid shirt gets excited about the game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 6.0], "base_se_norm": [-0.8118051290512085, -0.7741661667823792], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.24208605289459229, 0.317363977432251], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001121.pt"}
{"idx": 1123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRjCaSrmm2A.mp4", "duration": 53.13641666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the title of the movie 'DEAD BODY' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the title of the movie 'DEAD BODY' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 38.0], "base_se_norm": [0.35500288009643555, 0.43028080463409424], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [0.35500288009643555, 0.43028080463409424], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001123.pt"}
{"idx": 1125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L8qOb8mdTTU.mp4", "duration": 146.56, "prompt_text": "Please find the visual event described by the sentence 'When does the cat first notice the spider on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat first notice the spider on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.8908296823501587, -0.836244523525238], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.8498908281326294, -0.8225982189178467], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001125.pt"}
{"idx": 1127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L8qOb8mdTTU.mp4", "duration": 146.56, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cat is pulling clothes out of the laundry basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cat is pulling clothes out of the laundry basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 68.0], "base_se_norm": [-0.317685604095459, -0.07205235958099365], "gt_se_raw": [63.0, 67.0], "gt_se_norm": [-0.14028382301330566, -0.08569866418838501], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001127.pt"}
{"idx": 1129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L8qOb8mdTTU.mp4", "duration": 146.56, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cat gets stuck to the snowboard due to static electricity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cat gets stuck to the snowboard due to static electricity.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 115.0], "base_se_norm": [0.5147379636764526, 0.5693231821060181], "gt_se_raw": [112.0, 114.0], "gt_se_norm": [0.5283843278884888, 0.5556769371032715], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001129.pt"}
{"idx": 1131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HodFzcJHIYI.mp4", "duration": 188.48, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man told that he is blind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man told that he is blind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [-0.6498302221298218, -0.6179965734481812], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.6392190456390381, -0.6179965734481812], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001131.pt"}
{"idx": 1133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HodFzcJHIYI.mp4", "duration": 188.48, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a woman is teaching a blind man to use a cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a woman is teaching a blind man to use a cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 73.0], "base_se_norm": [-0.2890492081642151, -0.2253819704055786], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [-0.26782679557800293, -0.23599320650100708], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001133.pt"}
{"idx": 1135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-Kb1VHx2pKk.mp4", "duration": 171.03753333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman setting up the number 468 on the abacus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman setting up the number 468 on the abacus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 48.0], "base_se_norm": [-0.8713732957839966, -0.4387196898460388], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.953226625919342, -0.9181466102600098], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001135.pt"}
{"idx": 1137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-Kb1VHx2pKk.mp4", "duration": 171.03753333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the instructor demonstrate how to subtract 372 from 707?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the instructor demonstrate how to subtract 372 from 707?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 120.0], "base_se_norm": [0.05240058898925781, 0.40320074558258057], "gt_se_raw": [55.0, 79.0], "gt_se_norm": [-0.3568663001060486, -0.07622617483139038], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001137.pt"}
{"idx": 1139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-Kb1VHx2pKk.mp4", "duration": 171.03753333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the final problem, 843 minus 275, is being solved.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the final problem, 843 minus 275, is being solved.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 167.0], "base_se_norm": [0.5786008834838867, 0.9527877569198608], "gt_se_raw": [141.0, 162.0], "gt_se_norm": [0.6487609148025513, 0.8943209648132324], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001139.pt"}
{"idx": 1141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uceySVBjKNw.mp4", "duration": 89.52276666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy have his head down in the wagon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy have his head down in the wagon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 42.0], "base_se_norm": [-0.3968014717102051, -0.06169116497039795], "gt_se_raw": [26.0, 47.0], "gt_se_norm": [-0.4191421866416931, 0.05001223087310791], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001141.pt"}
{"idx": 1143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uceySVBjKNw.mp4", "duration": 89.52276666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the woman pulls the wagon across the green carpet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the woman pulls the wagon across the green carpet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 82.0], "base_se_norm": [0.6755514144897461, 0.831936240196228], "gt_se_raw": [76.0, 83.0], "gt_se_norm": [0.6978921890258789, 0.8542768955230713], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001143.pt"}
{"idx": 1145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h4QzRfvkJZ4.mp4", "duration": 157.69920833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video show an old, black and white picture of a church?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show an old, black and white picture of a church?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.6829407215118408, -0.6448935866355896], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.6829407215118408, -0.6575759649276733], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001145.pt"}
{"idx": 1147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h4QzRfvkJZ4.mp4", "duration": 157.69920833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the blue shirt is holding up different outfits.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the blue shirt is holding up different outfits.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 108.0], "base_se_norm": [0.039954543113708496, 0.3696962594985962], "gt_se_raw": [102.0, 109.0], "gt_se_norm": [0.29360198974609375, 0.38237857818603516], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001147.pt"}
{"idx": 1149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h4QzRfvkJZ4.mp4", "duration": 157.69920833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the tank top find and read the note?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the tank top find and read the note?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 149.0], "base_se_norm": [0.7501673698425293, 0.8896734714508057], "gt_se_raw": [135.0, 149.0], "gt_se_norm": [0.7121202945709229, 0.8896734714508057], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001149.pt"}
{"idx": 1151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dh1V8GyyNYE.mp4", "duration": 142.85104166666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man with his face covered in white powder looking up in surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man with his face covered in white powder looking up in surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.6919868588447571, -0.6219838857650757], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [-0.6919868588447571, -0.6359844207763672], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001151.pt"}
{"idx": 1153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Dh1V8GyyNYE.mp4", "duration": 142.85104166666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the first paint can hits one of the men in the face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the first paint can hits one of the men in the face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [-0.08996117115020752, -0.047959327697753906], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [-0.1039617657661438, -0.061959922313690186], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001153.pt"}
{"idx": 1155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzrlM6uX9oU.mp4", "duration": 93.72706039372706, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where two women are talking in an office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where two women are talking in an office.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 34.0], "base_se_norm": [-0.40252047777175903, -0.2744891047477722], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.38118189573287964, -0.2958276867866516], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001155.pt"}
{"idx": 1157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzrlM6uX9oU.mp4", "duration": 93.72706039372706, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman in sunglasses and a red shirt is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman in sunglasses and a red shirt is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 49.0], "base_se_norm": [-0.10378068685531616, 0.04558920860290527], "gt_se_raw": [42.0, 49.0], "gt_se_norm": [-0.10378068685531616, 0.04558920860290527], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001157.pt"}
{"idx": 1159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzrlM6uX9oU.mp4", "duration": 93.72706039372706, "prompt_text": "Please find the visual event described by the sentence 'Locate the group photo of people standing in front of a blue golf cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the group photo of people standing in front of a blue golf cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 83.0], "base_se_norm": [0.6857458353042603, 0.7711000442504883], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.6857458353042603, 0.7711000442504883], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001159.pt"}
{"idx": 1161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qtLVMNUqlLk.mp4", "duration": 189.314125, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the NNELS logo with red hexagons is displayed against a green background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the NNELS logo with red hexagons is displayed against a green background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.7570176124572754, -0.7147598266601562], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.7570176124572754, -0.7041953206062317], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001161.pt"}
{"idx": 1163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qtLVMNUqlLk.mp4", "duration": 189.314125, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a hand-drawn illustration of a browser window with 'https://' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a hand-drawn illustration of a browser window with 'https://' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.3978261947631836, -0.34500396251678467], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [-0.3872617483139038, -0.34500396251678467], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001163.pt"}
{"idx": 1165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tByqpujET9o.mp4", "duration": 394.0603333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in the blue suit introducing herself and guiding the other women into the shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in the blue suit introducing herself and guiding the other women into the shop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.949246346950531], "gt_se_raw": [20.0, 28.0], "gt_se_norm": [-0.898492693901062, -0.8578897714614868], "start_token_span": [81, 84], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001165.pt"}
{"idx": 1167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tByqpujET9o.mp4", "duration": 394.0603333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the polka-dot shirt lying on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the polka-dot shirt lying on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 116.0], "base_se_norm": [-0.44170987606048584, -0.41125768423080444], "gt_se_raw": [110.0, 116.0], "gt_se_norm": [-0.44170987606048584, -0.41125768423080444], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001167.pt"}
{"idx": 1169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tByqpujET9o.mp4", "duration": 394.0603333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the bride in the wedding dress run out of the store and sit down in the middle of the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 349.0 - 366.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bride in the wedding dress run out of the store and sit down in the middle of the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 349.0 - 366.0 seconds.<|im_end|>\n", "base_se_raw": [349.0, 366.0], "base_se_norm": [0.771302342414856, 0.8575835227966309], "gt_se_raw": [350.0, 367.0], "gt_se_norm": [0.7763776779174805, 0.8626589775085449], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001169.pt"}
{"idx": 1171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9iQX13HWUzk.mp4", "duration": 81.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the dog is looking up at the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the dog is looking up at the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 52.0], "base_se_norm": [0.06063711643218994, 0.2826310396194458], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [0.18396711349487305, 0.23329901695251465], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001171.pt"}
{"idx": 1173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9iQX13HWUzk.mp4", "duration": 81.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the dog licks its nose and then looks down guiltily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the dog licks its nose and then looks down guiltily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.6300102472305298, -0.48201441764831543], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.18602263927459717, -0.11202466487884521], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001173.pt"}
{"idx": 1175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hm_HXH2z-2M.mp4", "duration": 5.638966666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman stirring the food in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman stirring the food in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, 0.41869962215423584], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.2906501889228821], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001175.pt"}
{"idx": 1177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hm_HXH2z-2M.mp4", "duration": 5.638966666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is feeding the child with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is feeding the child with a spoon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [0.06402480602264404, 0.7733745574951172], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [0.06402480602264404, 0.7733745574951172], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001177.pt"}
{"idx": 1179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hm_HXH2z-2M.mp4", "duration": 5.638966666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the spoon enters the child's mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the spoon enters the child's mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [0.06402480602264404, 0.7733745574951172], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [0.41869962215423584, 0.7733745574951172], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001179.pt"}
{"idx": 1181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SxZgTzn64ME.mp4", "duration": 166.41625, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man making coffee using an espresso machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man making coffee using an espresso machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.7235846519470215, -0.675512433052063], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.7235846519470215, -0.675512433052063], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001181.pt"}
{"idx": 1183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SxZgTzn64ME.mp4", "duration": 166.41625, "prompt_text": "Please find the visual event described by the sentence 'Show the man purchasing and receiving food from a food truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man purchasing and receiving food from a food truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 121.0], "base_se_norm": [0.358040452003479, 0.45418500900268555], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [0.4181307554244995, 0.45418500900268555], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001183.pt"}
{"idx": 1185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PbrPhIn0cKk.mp4", "duration": 128.4283, "prompt_text": "Please find the visual event described by the sentence 'When are the four doctors displayed with their corresponding numbers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the four doctors displayed with their corresponding numbers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9532813429832458], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9377084374427795], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001185.pt"}
{"idx": 1187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PbrPhIn0cKk.mp4", "duration": 128.4283, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the 'Disability Discrimination Act 1992'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the 'Disability Discrimination Act 1992'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 59.0], "base_se_norm": [-0.3459385633468628, -0.08119940757751465], "gt_se_raw": [42.0, 59.0], "gt_se_norm": [-0.3459385633468628, -0.08119940757751465], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001187.pt"}
{"idx": 1189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PbrPhIn0cKk.mp4", "duration": 128.4283, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the white top is speaking directly to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the white top is speaking directly to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 128.0], "base_se_norm": [0.9310386180877686, 0.9933301210403442], "gt_se_raw": [124.0, 128.0], "gt_se_norm": [0.9310386180877686, 0.9933301210403442], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001189.pt"}
{"idx": 1191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_-O6Ppkrf98.mp4", "duration": 61.2, "prompt_text": "Please find the visual event described by the sentence 'When is the map showing the paths of the two men, one in red and one in purple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the map showing the paths of the two men, one in red and one in purple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.24836599826812744, -0.08496731519699097], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.24836599826812744, -0.08496731519699097], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001191.pt"}
{"idx": 1193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_-O6Ppkrf98.mp4", "duration": 61.2, "prompt_text": "Please find the visual event described by the sentence 'When do they use a glider to fly in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do they use a glider to fly in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 51.0], "base_se_norm": [0.4379085302352905, 0.6666666269302368], "gt_se_raw": [45.0, 50.0], "gt_se_norm": [0.47058820724487305, 0.6339869499206543], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001193.pt"}
{"idx": 1195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5n0EFtOuH9s.mp4", "duration": 179.77960000000002, "prompt_text": "Please find the visual event described by the sentence 'When is butter added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is butter added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.7775053381919861, -0.7441311478614807], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.7663806080818176, -0.7441311478614807], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001195.pt"}
{"idx": 1197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5n0EFtOuH9s.mp4", "duration": 179.77960000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point is the mixture stirred until it becomes thick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the mixture stirred until it becomes thick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 68.0], "base_se_norm": [-0.39926445484161377, -0.2435181736946106], "gt_se_raw": [54.0, 68.0], "gt_se_norm": [-0.39926445484161377, -0.2435181736946106], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001197.pt"}
{"idx": 1199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5n0EFtOuH9s.mp4", "duration": 179.77960000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a chocolate ball is being coated with crushed pistachios.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a chocolate ball is being coated with crushed pistachios.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 149.0], "base_se_norm": [0.5352131128311157, 0.6575851440429688], "gt_se_raw": [139.0, 149.0], "gt_se_norm": [0.5463378429412842, 0.6575851440429688], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001199.pt"}
{"idx": 1201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFQB8hJdLZw.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man guiding a patient with crutches up a flight of wooden stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man guiding a patient with crutches up a flight of wooden stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001201.pt"}
{"idx": 1203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFQB8hJdLZw.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a red shirt supervising another woman who is performing a leg exercise on a machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a red shirt supervising another woman who is performing a leg exercise on a machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001203.pt"}
{"idx": 1205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3LrBM_X-CYc.mp4", "duration": 23.967, "prompt_text": "Please find the visual event described by the sentence 'When do the colorful balloons begin to fall from the ceiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the colorful balloons begin to fall from the ceiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.6662076711654663, -0.3324154019355774], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.6662076711654663, -0.4158634543418884], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001205.pt"}
{"idx": 1207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3LrBM_X-CYc.mp4", "duration": 23.967, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a girl in a pink dress is jumping and playing among the balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a girl in a pink dress is jumping and playing among the balloons.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.3324154019355774, 0.2517211437225342], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.3324154019355774, 0.0013769865036010742], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001207.pt"}
{"idx": 1209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3LrBM_X-CYc.mp4", "duration": 23.967, "prompt_text": "Please find the visual event described by the sentence 'When is a woman on the stairs seen taking a picture with her camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman on the stairs seen taking a picture with her camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 23.0], "base_se_norm": [0.3351691961288452, 0.9193058013916016], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [0.2517211437225342, 0.5020654201507568], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001209.pt"}
{"idx": 1211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t7Vtyztl4ro.mp4", "duration": 34.134100000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy moves the walker over the second striped bar on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy moves the walker over the second striped bar on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [0.054663777351379395, 0.5234031677246094], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.3554832339286804, -0.2382984161376953], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001211.pt"}
{"idx": 1213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t7Vtyztl4ro.mp4", "duration": 34.134100000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the boy push off with his right cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy push off with his right cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.41407567262649536, -0.2382984161376953], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.28903353214263916, 0.4062182903289795], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001213.pt"}
{"idx": 1215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bMz-xG9gk8k.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the teacher and student are using the large vertical board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the teacher and student are using the large vertical board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, -0.6336996555328369], "gt_se_raw": [0.0, 32.0], "gt_se_norm": [-1.0, -0.46719950437545776], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001215.pt"}
{"idx": 1217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bMz-xG9gk8k.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher sign the word 'explore' into the student's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher sign the word 'explore' into the student's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 58.0], "base_se_norm": [-0.46719950437545776, -0.03429907560348511], "gt_se_raw": [41.0, 42.0], "gt_se_norm": [-0.31734931468963623, -0.30069929361343384], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001217.pt"}
{"idx": 1219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bMz-xG9gk8k.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Show me when the student is learning about 'choice time'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the student is learning about 'choice time'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 90.0], "base_se_norm": [-0.46719950437545776, 0.49850142002105713], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.6816515922546387, 0.7149516344070435], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001219.pt"}
{"idx": 1221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CSlsvxgfhE.mp4", "duration": 439.50573333333335, "prompt_text": "Please find the visual event described by the sentence 'At what time do the man and woman in the red dress have a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do the man and woman in the red dress have a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 153.0], "base_se_norm": [-0.7861233949661255, -0.30376332998275757], "gt_se_raw": [47.0, 153.0], "gt_se_norm": [-0.7861233949661255, -0.30376332998275757], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001221.pt"}
{"idx": 1223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CSlsvxgfhE.mp4", "duration": 439.50573333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the park fall down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 357.0 - 360.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the park fall down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 357.0 - 360.0 seconds.<|im_end|>\n", "base_se_raw": [357.0, 360.0], "base_se_norm": [0.6245521306991577, 0.6382038593292236], "gt_se_raw": [357.0, 360.0], "gt_se_norm": [0.6245521306991577, 0.6382038593292236], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001223.pt"}
{"idx": 1225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdZMypElBpo.mp4", "duration": 61.60320833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the child dressed as Darth Vader tries to use the force on an exercise bike.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the child dressed as Darth Vader tries to use the force on an exercise bike.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.8052049279212952, -0.7078074216842651], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.8052049279212952, -0.7078074216842651], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001225.pt"}
{"idx": 1227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdZMypElBpo.mp4", "duration": 61.60320833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman place a sandwich on the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman place a sandwich on the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [-0.18835395574569702, -0.12342226505279541], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.15588808059692383, -0.09095638990402222], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001227.pt"}
{"idx": 1229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdZMypElBpo.mp4", "duration": 61.60320833333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where the man remotely starts the car using the key fob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where the man remotely starts the car using the key fob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [0.5908262729644775, 0.6882238388061523], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [0.5583604574203491, 0.623292088508606], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001229.pt"}
{"idx": 1231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gq0jfzgvvrM.mp4", "duration": 146.51303333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the big bird lands on the wire, causing it to sag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the big bird lands on the wire, causing it to sag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.4403228163719177, -0.39937084913253784], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.4403228163719177, -0.4130215048789978], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001231.pt"}
{"idx": 1233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gq0jfzgvvrM.mp4", "duration": 146.51303333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the small birds are pecking the big bird's feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the small birds are pecking the big bird's feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 128.0], "base_se_norm": [0.6517301797866821, 0.7472848892211914], "gt_se_raw": [76.0, 87.0], "gt_se_norm": [0.037450432777404785, 0.18760764598846436], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001233.pt"}
{"idx": 1235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WZ89Bgcdr1M.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man soldering copper pipes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man soldering copper pipes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001235.pt"}
{"idx": 1237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WZ89Bgcdr1M.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point are two men seen working on plumbing equipment together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two men seen working on plumbing equipment together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001237.pt"}
{"idx": 1239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WZ89Bgcdr1M.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where information about career opportunities is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where information about career opportunities is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001239.pt"}
{"idx": 1241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bLX3hv7cTOE.mp4", "duration": 154.4543, "prompt_text": "Please find the visual event described by the sentence 'At what point does the presenter explain the low-vision Webster-Pak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the presenter explain the low-vision Webster-Pak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 76.0], "base_se_norm": [-0.5597403049468994, -0.015890121459960938], "gt_se_raw": [81.0, 99.0], "gt_se_norm": [0.04885387420654297, 0.2819325923919678], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001241.pt"}
{"idx": 1243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bLX3hv7cTOE.mp4", "duration": 154.4543, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the presenter demonstrates how to use the blue tool to get the pills out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the presenter demonstrates how to use the blue tool to get the pills out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 125.0], "base_se_norm": [0.5279600620269775, 0.6186017990112305], "gt_se_raw": [117.0, 125.0], "gt_se_norm": [0.515011191368103, 0.6186017990112305], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001243.pt"}
{"idx": 1245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z0yFqQWLa0M.mp4", "duration": 84.37595555555556, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman in the floral jacket folds her arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman in the floral jacket folds her arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.6918553113937378, -0.620745062828064], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.5496346950531006, -0.43111753463745117], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001245.pt"}
{"idx": 1247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z0yFqQWLa0M.mp4", "duration": 84.37595555555556, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman says she should get going and turns to leave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman says she should get going and turns to leave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [0.5644267797470093, 0.6829439401626587], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [0.13776493072509766, 0.20887517929077148], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001247.pt"}
{"idx": 1249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z0yFqQWLa0M.mp4", "duration": 84.37595555555556, "prompt_text": "Please find the visual event described by the sentence 'When does the woman rush back to the man and they begin to kiss passionately?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman rush back to the man and they begin to kiss passionately?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 84.0], "base_se_norm": [0.8014611005783081, 0.9910886287689209], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [0.8251645565032959, 0.9199782609939575], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001249.pt"}
{"idx": 1251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/db3Ep-jM6ZE.mp4", "duration": 17.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman places an object on the countertop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman places an object on the countertop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 13.0], "base_se_norm": [-0.31947070360183716, 0.47448015213012695], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.31947070360183716, -0.09262758493423462], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001251.pt"}
{"idx": 1253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/db3Ep-jM6ZE.mp4", "duration": 17.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the woman picks up the object and leaves the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the woman picks up the object and leaves the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 15.0], "base_se_norm": [-0.5463137626647949, 0.7013232707977295], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [0.5879017114639282, 0.8147448301315308], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001253.pt"}
{"idx": 1255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jMsTvH6gTGk.mp4", "duration": 184.6, "prompt_text": "Please find the visual event described by the sentence 'When are examples of patterns in nature, music and clothing presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are examples of patterns in nature, music and clothing presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 24.0], "base_se_norm": [-0.9349945783615112, -0.7399783134460449], "gt_se_raw": [7.0, 24.0], "gt_se_norm": [-0.924160361289978, -0.7399783134460449], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001255.pt"}
{"idx": 1257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jMsTvH6gTGk.mp4", "duration": 184.6, "prompt_text": "Please find the visual event described by the sentence 'Show the part where an ABC pattern is demonstrated with a fork, a spoon, and a mug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where an ABC pattern is demonstrated with a fork, a spoon, and a mug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 101.0], "base_se_norm": [-0.3174431324005127, 0.09425783157348633], "gt_se_raw": [63.0, 81.0], "gt_se_norm": [-0.3174431324005127, -0.12242692708969116], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001257.pt"}
{"idx": 1259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jMsTvH6gTGk.mp4", "duration": 184.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a pattern with leaves and shells is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a pattern with leaves and shells is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 139.0], "base_se_norm": [0.24593710899353027, 0.5059587955474854], "gt_se_raw": [116.0, 128.0], "gt_se_norm": [0.2567713260650635, 0.386782169342041], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001259.pt"}
{"idx": 1261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QNJx7Vi4Sg4.mp4", "duration": 133.80033333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the diving suit emerge from the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the diving suit emerge from the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 55.0], "base_se_norm": [-0.297460675239563, -0.17787951231002808], "gt_se_raw": [47.0, 56.0], "gt_se_norm": [-0.297460675239563, -0.16293185949325562], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001261.pt"}
{"idx": 1263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QNJx7Vi4Sg4.mp4", "duration": 133.80033333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the whole family share a group hug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the whole family share a group hug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 115.0], "base_se_norm": [0.3751833438873291, 0.7189792394638062], "gt_se_raw": [92.0, 107.0], "gt_se_norm": [0.3751833438873291, 0.5993980169296265], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001263.pt"}
{"idx": 1265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vm1Zeu-DpmU.mp4", "duration": 201.70149999999998, "prompt_text": "Please find the visual event described by the sentence 'When are the purple grape-flavored candies being sorted on the conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the purple grape-flavored candies being sorted on the conveyor belt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.970253050327301], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9603374004364014], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001265.pt"}
{"idx": 1267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vm1Zeu-DpmU.mp4", "duration": 201.70149999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dark fruit concentrate added to the white candy base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dark fruit concentrate added to the white candy base?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 98.0], "base_se_norm": [-0.09767657518386841, -0.0282670259475708], "gt_se_raw": [91.0, 98.0], "gt_se_norm": [-0.09767657518386841, -0.0282670259475708], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001267.pt"}
{"idx": 1269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vm1Zeu-DpmU.mp4", "duration": 201.70149999999998, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the baseball team logos are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 170.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the baseball team logos are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 170.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 170.0], "base_se_norm": [0.6559122800827026, 0.6856591701507568], "gt_se_raw": [167.0, 170.0], "gt_se_norm": [0.6559122800827026, 0.6856591701507568], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001269.pt"}
{"idx": 1271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OpcxenhBO64.mp4", "duration": 195.2951, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the paper is folded to form a boat-like shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the paper is folded to form a boat-like shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [-0.7439771890640259, -0.6108453273773193], "gt_se_raw": [27.0, 38.0], "gt_se_norm": [-0.723495364189148, -0.6108453273773193], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001271.pt"}
{"idx": 1273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OpcxenhBO64.mp4", "duration": 195.2951, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the paper is folded like an accordion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the paper is folded like an accordion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 128.0], "base_se_norm": [-0.3650634288787842, 0.3108367919921875], "gt_se_raw": [109.0, 123.0], "gt_se_norm": [0.11625945568084717, 0.2596322298049927], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001273.pt"}
{"idx": 1275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPbvr9LH84w.mp4", "duration": 184.184, "prompt_text": "Please find the visual event described by the sentence 'When are the cheerleaders performing their dance routine in the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the cheerleaders performing their dance routine in the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 89.0], "base_se_norm": [-0.196455717086792, -0.03357517719268799], "gt_se_raw": [80.0, 91.0], "gt_se_norm": [-0.13130348920822144, -0.011857748031616211], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001275.pt"}
{"idx": 1277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPbvr9LH84w.mp4", "duration": 184.184, "prompt_text": "Please find the visual event described by the sentence 'At what point is the red-haired man talking to a man in a grey tank top in the yard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the red-haired man talking to a man in a grey tank top in the yard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 30.0], "base_se_norm": [-0.88055419921875, -0.6742388010025024], "gt_se_raw": [16.0, 30.0], "gt_se_norm": [-0.8262606859207153, -0.6742388010025024], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001277.pt"}
{"idx": 1279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GPbvr9LH84w.mp4", "duration": 184.184, "prompt_text": "Please find the visual event described by the sentence 'When are the two women having a conversation with another woman who is pouring orange juice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women having a conversation with another woman who is pouring orange juice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 63.0], "base_se_norm": [-0.48964083194732666, -0.315901517868042], "gt_se_raw": [47.0, 63.0], "gt_se_norm": [-0.48964083194732666, -0.315901517868042], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001279.pt"}
{"idx": 1281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9jX0S7mvB8.mp4", "duration": 151.6515, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man kiss the woman on the cheek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man kiss the woman on the cheek?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.7626136541366577, -0.6966729760169983], "gt_se_raw": [18.0, 23.0], "gt_se_norm": [-0.7626136541366577, -0.6966729760169983], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001281.pt"}
{"idx": 1283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9jX0S7mvB8.mp4", "duration": 151.6515, "prompt_text": "Please find the visual event described by the sentence 'When are the man and woman having an argument on the cobblestone street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and woman having an argument on the cobblestone street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 121.0], "base_se_norm": [-0.5252272486686707, 0.5957639217376709], "gt_se_raw": [64.0, 102.0], "gt_se_norm": [-0.1559596061706543, 0.3451894521713257], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001283.pt"}
{"idx": 1285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-wHytb5Fe2k.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'When does the woman guide the boy to find the letter 'V'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman guide the boy to find the letter 'V'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 18.0], "base_se_norm": [-0.9626168012619019, -0.32710278034210205], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.7383177280426025, -0.5140186548233032], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001285.pt"}
{"idx": 1287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-wHytb5Fe2k.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman praise the boy by saying 'You're finished' and clapping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman praise the boy by saying 'You're finished' and clapping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.32710278034210205, -0.17757010459899902], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.2897196412086487, -0.17757010459899902], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001287.pt"}
{"idx": 1289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-wHytb5Fe2k.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the boy is happily playing with the blue spinning toy he received.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the boy is happily playing with the blue spinning toy he received.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 53.0], "base_se_norm": [0.6822429895401001, 0.9813084602355957], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [0.6822429895401001, 0.9439252614974976], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001289.pt"}
{"idx": 1291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uWuOBonC1ds.mp4", "duration": 312.37873333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy put a gallon of milk into his shopping cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy put a gallon of milk into his shopping cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [-0.5262161493301392, -0.5006061792373657], "gt_se_raw": [74.0, 78.0], "gt_se_norm": [-0.5262161493301392, -0.5006061792373657], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001291.pt"}
{"idx": 1293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uWuOBonC1ds.mp4", "duration": 312.37873333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the little girl is pushing the toddler who is sitting inside a shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 190.0 - 200.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the little girl is pushing the toddler who is sitting inside a shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 190.0 - 200.0 seconds.<|im_end|>\n", "base_se_raw": [190.0, 200.0], "base_se_norm": [0.21647214889526367, 0.2804969549179077], "gt_se_raw": [191.0, 199.0], "gt_se_norm": [0.22287452220916748, 0.27409446239471436], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001293.pt"}
{"idx": 1295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PG47b8hd2o.mp4", "duration": 85.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker first mention the blended learning planning document?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker first mention the blended learning planning document?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9765899181365967, -0.7658993601799011], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8361295461654663, -0.7893093824386597], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001295.pt"}
{"idx": 1297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PG47b8hd2o.mp4", "duration": 85.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where he states that online teaching is not radically different from face-to-face teaching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where he states that online teaching is not radically different from face-to-face teaching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 16.0], "base_se_norm": [-0.7658993601799011, -0.625438928604126], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [0.05345296859741211, 0.12368321418762207], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001297.pt"}
{"idx": 1299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PG47b8hd2o.mp4", "duration": 85.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention that students' learning will also improve?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention that students' learning will also improve?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [0.17050325870513916, 0.24073350429534912], "gt_se_raw": [71.0, 73.0], "gt_se_norm": [0.6621147394180298, 0.7089347839355469], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001299.pt"}
{"idx": 1301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Nb8KXFH3aMs.mp4", "duration": 54.68, "prompt_text": "Please find the visual event described by the sentence 'Show me when a dog is running fast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a dog is running fast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.8171178102493286, -0.561082661151886], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.8171178102493286, -0.7073884010314941], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001301.pt"}
{"idx": 1303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Nb8KXFH3aMs.mp4", "duration": 54.68, "prompt_text": "Please find the visual event described by the sentence 'When are the Golden Retriever and the brown dog seen playing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the Golden Retriever and the brown dog seen playing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.30504757165908813, -0.23189467191696167], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.30504757165908813, -0.23189467191696167], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001303.pt"}
{"idx": 1305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgg9jE1jXjg.mp4", "duration": 161.69486666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the football players eating yogurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the football players eating yogurt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.616561770439148, -0.579454779624939], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.6289306879043579, -0.5918238162994385], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001305.pt"}
{"idx": 1307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgg9jE1jXjg.mp4", "duration": 161.69486666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a group of football players is excitedly dancing and jumping in a tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a group of football players is excitedly dancing and jumping in a tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 71.0], "base_se_norm": [-0.17127859592437744, -0.1218026876449585], "gt_se_raw": [67.0, 71.0], "gt_se_norm": [-0.17127859592437744, -0.1218026876449585], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001307.pt"}
{"idx": 1309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgg9jE1jXjg.mp4", "duration": 161.69486666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the drill where a woman in a white t-shirt pushes a tackling dummy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the drill where a woman in a white t-shirt pushes a tackling dummy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [0.3482184410095215, 0.39769434928894043], "gt_se_raw": [109.0, 112.0], "gt_se_norm": [0.3482184410095215, 0.3853253126144409], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001309.pt"}
{"idx": 1311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zfn8Mtyu5s0.mp4", "duration": 113.9, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the yarn is first wound into the notch of the nostepinne.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the yarn is first wound into the notch of the nostepinne.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 46.0], "base_se_norm": [-0.5610184669494629, -0.19227391481399536], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.5434591770172119, -0.47322213649749756], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001311.pt"}
{"idx": 1313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zfn8Mtyu5s0.mp4", "duration": 113.9, "prompt_text": "Please find the visual event described by the sentence 'When can we see the ball of yarn growing into an egg shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the ball of yarn growing into an egg shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 97.0], "base_se_norm": [0.08867430686950684, 0.7032483816146851], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [0.1589113473892212, 0.22914838790893555], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001313.pt"}
{"idx": 1315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSfeGBrchjU.mp4", "duration": 110.56, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the green jacket running away from the children?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the green jacket running away from the children?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 24.0], "base_se_norm": [-0.8010129928588867, -0.5658465623855591], "gt_se_raw": [8.0, 24.0], "gt_se_norm": [-0.855282187461853, -0.5658465623855591], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001315.pt"}
{"idx": 1317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSfeGBrchjU.mp4", "duration": 110.56, "prompt_text": "Please find the visual event described by the sentence 'Show me when the children wearing red hats are running around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the children wearing red hats are running around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.3668596148490906, -0.25832128524780273], "gt_se_raw": [35.0, 42.0], "gt_se_norm": [-0.3668596148490906, -0.24023151397705078], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001317.pt"}
{"idx": 1319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSfeGBrchjU.mp4", "duration": 110.56, "prompt_text": "Please find the visual event described by the sentence 'Locate the time a girl in an orange top is pulled away from the line of children.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time a girl in an orange top is pulled away from the line of children.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 84.0], "base_se_norm": [0.4471780061721802, 0.5195369720458984], "gt_se_raw": [77.0, 83.0], "gt_se_norm": [0.39290881156921387, 0.5014472007751465], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001319.pt"}
{"idx": 1321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3nYvJY2C9B4.mp4", "duration": 86.01926666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cat meows towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cat meows towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.581489086151123, -0.488486647605896], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.6047396659851074, -0.488486647605896], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001321.pt"}
{"idx": 1323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3nYvJY2C9B4.mp4", "duration": 86.01926666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the cat starts to eat the yogurt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the cat starts to eat the yogurt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 56.0], "base_se_norm": [0.023026704788208008, 0.3020339012145996], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [0.04627728462219238, 0.20903146266937256], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001323.pt"}
{"idx": 1325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ehJfUqSbPtg.mp4", "duration": 88.09693031441432, "prompt_text": "Please find the visual event described by the sentence 'When does a cat emerge from the play tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a cat emerge from the play tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9318932294845581], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9545954465866089], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001325.pt"}
{"idx": 1327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ehJfUqSbPtg.mp4", "duration": 88.09693031441432, "prompt_text": "Please find the visual event described by the sentence 'Show me when there are three cats together watching the robot vacuum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when there are three cats together watching the robot vacuum.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 38.0], "base_se_norm": [-0.4778478741645813, -0.13731390237808228], "gt_se_raw": [25.0, 39.0], "gt_se_norm": [-0.4324433207511902, -0.11461162567138672], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001327.pt"}
{"idx": 1329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ehJfUqSbPtg.mp4", "duration": 88.09693031441432, "prompt_text": "Please find the visual event described by the sentence 'When does the robot vacuum move towards the cat sitting at the top of the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the robot vacuum move towards the cat sitting at the top of the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [0.29402923583984375, 0.43024277687072754], "gt_se_raw": [70.0, 75.0], "gt_se_norm": [0.5891586542129517, 0.7026699781417847], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001329.pt"}
{"idx": 1331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIb_741_dIw.mp4", "duration": 89.214125, "prompt_text": "Please find the visual event described by the sentence 'When are the eggs added to the butter and sugar mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the eggs added to the butter and sugar mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.6637303829193115, -0.574058473110199], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.6637303829193115, -0.5964764356613159], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001331.pt"}
{"idx": 1333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lIb_741_dIw.mp4", "duration": 89.214125, "prompt_text": "Please find the visual event described by the sentence 'At what point is the brownie batter poured and spread into the baking pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the brownie batter poured and spread into the baking pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 53.0], "base_se_norm": [0.03122687339782715, 0.18815267086029053], "gt_se_raw": [46.0, 53.0], "gt_se_norm": [0.03122687339782715, 0.18815267086029053], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001333.pt"}
{"idx": 1335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJnpwL-2Drc.mp4", "duration": 142.28, "prompt_text": "Please find the visual event described by the sentence 'When does the baby suddenly become an adult?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby suddenly become an adult?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [-0.6345235109329224, -0.592353105545044], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.6204667091369629, -0.592353105545044], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001335.pt"}
{"idx": 1337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJnpwL-2Drc.mp4", "duration": 142.28, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the patient shoot the doctor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the patient shoot the doctor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 112.0], "base_se_norm": [0.5321900844573975, 0.5743603706359863], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.1847062110900879, -0.15659260749816895], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001337.pt"}
{"idx": 1339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJnpwL-2Drc.mp4", "duration": 142.28, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a person steals a police car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a person steals a police car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [-0.1144222617149353, -0.05819511413574219], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.08630865812301636, -0.044138312339782715], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001339.pt"}
{"idx": 1341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VGb1wgIkPEw.mp4", "duration": 11.541666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the red fire truck drives across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the red fire truck drives across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.30685919523239136, 0.3862816095352173], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.30685919523239136, 0.21299636363983154], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001341.pt"}
{"idx": 1343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VGb1wgIkPEw.mp4", "duration": 11.541666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the camera's perspective is shifting from the parked car to the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the camera's perspective is shifting from the parked car to the road.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.4801444411277771, 0.0397111177444458], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.4801444411277771, 0.0397111177444458], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001343.pt"}
{"idx": 1345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkTpA2U4ooA.mp4", "duration": 150.98416666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see the player in blue armor falling into the trap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the player in blue armor falling into the trap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8542893528938293, -0.7880572080612183], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.8410429358482361, -0.8013036251068115], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001345.pt"}
{"idx": 1347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkTpA2U4ooA.mp4", "duration": 150.98416666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the green team's bed is destroyed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the green team's bed is destroyed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5363752245903015, -0.49663591384887695], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.5363752245903015, -0.5098823308944702], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001347.pt"}
{"idx": 1349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dkTpA2U4ooA.mp4", "duration": 150.98416666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the 'VICTORY!' message appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the 'VICTORY!' message appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 138.0], "base_se_norm": [0.7882670164108276, 0.8280062675476074], "gt_se_raw": [136.0, 138.0], "gt_se_norm": [0.8015134334564209, 0.8280062675476074], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001349.pt"}
{"idx": 1351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-tH4HbncWNs.mp4", "duration": 122.49737499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Mickey and Pluto are about to eat the strawberry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Mickey and Pluto are about to eat the strawberry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 64.0], "base_se_norm": [-0.06936782598495483, 0.04492032527923584], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [-0.004060268402099609, 0.04492032527923584], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001351.pt"}
{"idx": 1353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-tH4HbncWNs.mp4", "duration": 122.49737499999999, "prompt_text": "Please find the visual event described by the sentence 'At what time do all the Tsum Tsums gather for a group photo around the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do all the Tsum Tsums gather for a group photo around the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [0.7796299457550049, 0.8449374437332153], "gt_se_raw": [111.0, 115.0], "gt_se_norm": [0.8122837543487549, 0.8775912523269653], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001353.pt"}
{"idx": 1355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/beECZjCRLmQ.mp4", "duration": 33.44, "prompt_text": "Please find the visual event described by the sentence 'When does the person show the salt and pepper pots while they are still in the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person show the salt and pepper pots while they are still in the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.940191388130188, -0.5813397169113159], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.940191388130188, -0.7009569406509399], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001355.pt"}
{"idx": 1357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/beECZjCRLmQ.mp4", "duration": 33.44, "prompt_text": "Please find the visual event described by the sentence 'At what point is the second pot, the salt pot, removed from the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the second pot, the salt pot, removed from the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.4019138216972351, -0.2822965979576111], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.4617224931716919, -0.3421052098274231], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001357.pt"}
{"idx": 1359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/beECZjCRLmQ.mp4", "duration": 33.44, "prompt_text": "Please find the visual event described by the sentence 'When does a finger point to the pot that has no holes on top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a finger point to the pot that has no holes on top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.3755981922149658, 0.5550240278244019], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [0.3755981922149658, 0.5550240278244019], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001359.pt"}
{"idx": 1361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QmJ3xuJrUcM.mp4", "duration": 146.47966666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the crow is walking on the table, examining the setup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the crow is walking on the table, examining the setup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 74.0], "base_se_norm": [-0.19442743062973022, 0.010379195213317871], "gt_se_raw": [58.0, 69.0], "gt_se_norm": [-0.2080811858177185, -0.057889699935913086], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001361.pt"}
{"idx": 1363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QmJ3xuJrUcM.mp4", "duration": 146.47966666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the crow uses the short stick to pull the long stick out of the cage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the crow uses the short stick to pull the long stick out of the cage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 125.0], "base_se_norm": [0.5974913835525513, 0.7067215442657471], "gt_se_raw": [116.0, 125.0], "gt_se_norm": [0.583837628364563, 0.7067215442657471], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001363.pt"}
{"idx": 1365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zMB-bVwEi2Q.mp4", "duration": 72.33893333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the boy clapping his hands in excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy clapping his hands in excitement?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.28116166591644287, -0.225866436958313], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [-0.28116166591644287, -0.25351405143737793], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001365.pt"}
{"idx": 1367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zMB-bVwEi2Q.mp4", "duration": 72.33893333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman place her hand on the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman place her hand on the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.6958761215209961, -0.6129332184791565], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [0.13355278968811035, 0.18884801864624023], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001367.pt"}
{"idx": 1369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zMB-bVwEi2Q.mp4", "duration": 72.33893333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy turn the paper feed knob?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy turn the paper feed knob?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.6682284474372864, -0.5852855443954468], "gt_se_raw": [69.0, 70.0], "gt_se_norm": [0.9076863527297974, 0.9353339672088623], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001369.pt"}
{"idx": 1371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mTLO2F_ERY.mp4", "duration": 123.84, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman's dress get blown up by the wind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman's dress get blown up by the wind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 25.0], "base_se_norm": [-0.7093023061752319, -0.5962532162666321], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8869509100914001, -0.8385012745857239], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001371.pt"}
{"idx": 1373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2mTLO2F_ERY.mp4", "duration": 123.84, "prompt_text": "Please find the visual event described by the sentence 'When does the man's wind slam the shutters closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man's wind slam the shutters closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.38630491495132446, -0.3378552794456482], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [-0.3217054009437561, -0.2894056439399719], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001373.pt"}
{"idx": 1375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R93ce4FZGbc.mp4", "duration": 116.116, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video are the sharks hunting the fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video are the sharks hunting the fish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 84.0], "base_se_norm": [0.18846678733825684, 0.44682908058166504], "gt_se_raw": [69.0, 77.0], "gt_se_norm": [0.18846678733825684, 0.326259970664978], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001375.pt"}
{"idx": 1377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R93ce4FZGbc.mp4", "duration": 116.116, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the green Grandpa Shark is featured.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the green Grandpa Shark is featured.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 68.0], "base_se_norm": [0.050673484802246094, 0.1712425947189331], "gt_se_raw": [61.0, 69.0], "gt_se_norm": [0.050673484802246094, 0.18846678733825684], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001377.pt"}
{"idx": 1379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dKXpAC311Lg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When is the bar chart that compares employment rates between two groups in 1991 and 2010 displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the bar chart that compares employment rates between two groups in 1991 and 2010 displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [86, 89], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001379.pt"}
{"idx": 1381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dKXpAC311Lg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the period where the 41.1% employment rate for disabled individuals in 2010 is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period where the 41.1% employment rate for disabled individuals in 2010 is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [86, 89], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001381.pt"}
{"idx": 1383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dKXpAC311Lg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When can we see the data corresponding to the year 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the data corresponding to the year 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001383.pt"}
{"idx": 1385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKDyFgUqWfQ.mp4", "duration": 224.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the player in cyan armor builds a bridge using blue blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the player in cyan armor builds a bridge using blue blocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.8839976191520691, -0.7501487135887146], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.9018441438674927, -0.8661510944366455], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001385.pt"}
{"idx": 1387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKDyFgUqWfQ.mp4", "duration": 224.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the green-clad player realize they are trapped and express their despair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the green-clad player realize they are trapped and express their despair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 78.0], "base_se_norm": [-0.3842950463294983, -0.30398571491241455], "gt_se_raw": [72.0, 76.0], "gt_se_norm": [-0.3575252890586853, -0.32183223962783813], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001387.pt"}
{"idx": 1389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pU0GSbe6r_4.mp4", "duration": 145.92, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Tom is preparing the poisoned milk for the mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Tom is preparing the poisoned milk for the mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.6847587823867798], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9451754093170166], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001389.pt"}
{"idx": 1391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pU0GSbe6r_4.mp4", "duration": 145.92, "prompt_text": "Please find the visual event described by the sentence 'When does Jerry drink from the bowl and then faint on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jerry drink from the bowl and then faint on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 52.0], "base_se_norm": [-0.3969298005104065, -0.2872806787490845], "gt_se_raw": [45.0, 51.0], "gt_se_norm": [-0.38322365283966064, -0.3009868264198303], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001391.pt"}
{"idx": 1393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pU0GSbe6r_4.mp4", "duration": 145.92, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the muscular Jerry pulls Tom out of the safe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the muscular Jerry pulls Tom out of the safe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 112.0], "base_se_norm": [0.43914473056793213, 0.5350877046585083], "gt_se_raw": [104.0, 107.0], "gt_se_norm": [0.42543864250183105, 0.46655702590942383], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001393.pt"}
{"idx": 1395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1yJA_vE6t9E.mp4", "duration": 166.69986666666668, "prompt_text": "Please find the visual event described by the sentence 'A boy in a blue t-shirt is using a tablet with an adult's help.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A boy in a blue t-shirt is using a tablet with an adult's help.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 106.0], "base_se_norm": [0.13977301120758057, 0.27174675464630127], "gt_se_raw": [95.0, 106.0], "gt_se_norm": [0.13977301120758057, 0.27174675464630127], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001395.pt"}
{"idx": 1397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1yJA_vE6t9E.mp4", "duration": 166.69986666666668, "prompt_text": "Please find the visual event described by the sentence 'A woman is drawing a car on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is drawing a car on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 136.0], "base_se_norm": [0.4277156591415405, 0.6316750049591064], "gt_se_raw": [123.0, 136.0], "gt_se_norm": [0.4757061004638672, 0.6316750049591064], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001397.pt"}
{"idx": 1399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpBW-5BtPoU.mp4", "duration": 124.79133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man explain that the theremin is a monophonic instrument?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man explain that the theremin is a monophonic instrument?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.5993311405181885, -0.519197404384613], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.5672776699066162, -0.5031706094741821], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001399.pt"}
{"idx": 1401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpBW-5BtPoU.mp4", "duration": 124.79133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man demonstrate changing the volume of the theremin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man demonstrate changing the volume of the theremin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 40.0], "base_se_norm": [-0.519197404384613, -0.3589298725128174], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [-0.27879607677459717, -0.1666088104248047], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001401.pt"}
{"idx": 1403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpBW-5BtPoU.mp4", "duration": 124.79133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the animated trivia section about the theremin's use in films and music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the animated trivia section about the theremin's use in films and music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 77.0], "base_se_norm": [-0.022368013858795166, 0.23406004905700684], "gt_se_raw": [62.0, 77.0], "gt_se_norm": [-0.006341278553009033, 0.23406004905700684], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001403.pt"}
{"idx": 1405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OSKNiwkaRRo.mp4", "duration": 19.5195, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in black communicates with both hands to the other person.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in black communicates with both hands to the other person.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [0.024616360664367676, 0.33200132846832275], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.8975383639335632, -0.5901534557342529], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001405.pt"}
{"idx": 1407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OSKNiwkaRRo.mp4", "duration": 19.5195, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person in yellow put their hand on the woman's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person in yellow put their hand on the woman's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.12707805633544922, 0.33200132846832275], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.12707805633544922, 0.33200132846832275], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001407.pt"}
{"idx": 1409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0P1Td5OTS-A.mp4", "duration": 40.04, "prompt_text": "Please find the visual event described by the sentence 'Where does the video compare the features of the Desktop and Online versions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where does the video compare the features of the Desktop and Online versions?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.7502497434616089, -0.5004994869232178], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.7502497434616089, -0.550449550151825], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001409.pt"}
{"idx": 1411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0P1Td5OTS-A.mp4", "duration": 40.04, "prompt_text": "Please find the visual event described by the sentence 'When are keyboard shortcuts mentioned as a feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are keyboard shortcuts mentioned as a feature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [0.048951029777526855, 0.1488511562347412], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [0.048951029777526855, 0.1488511562347412], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001411.pt"}
{"idx": 1413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0P1Td5OTS-A.mp4", "duration": 40.04, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the video encourages viewers to download the app.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the video encourages viewers to download the app.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [0.7482516765594482, 0.948051929473877], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [0.7482516765594482, 0.898101806640625], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001413.pt"}
{"idx": 1415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4A9lG59Tosg.mp4", "duration": 63.60520833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the hammer make contact with the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the hammer make contact with the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [-0.2139008641242981, -0.15101295709609985], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.30823278427124023, -0.24534481763839722], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001415.pt"}
{"idx": 1417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4A9lG59Tosg.mp4", "duration": 63.60520833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man's face appears over the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man's face appears over the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 42.0], "base_se_norm": [-0.025237083435058594, 0.3206465244293213], "gt_se_raw": [31.0, 42.0], "gt_se_norm": [-0.025237083435058594, 0.3206465244293213], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001417.pt"}
{"idx": 1419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5iu2FhCiQFw.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher present both the comb and the present to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher present both the comb and the present to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 26.0], "base_se_norm": [-0.8326735496520996, -0.5166124105453491], "gt_se_raw": [11.0, 18.0], "gt_se_norm": [-0.7954899072647095, -0.6653470993041992], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001419.pt"}
{"idx": 1421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5iu2FhCiQFw.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher use the red mitten to touch the girl's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher use the red mitten to touch the girl's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [0.059734344482421875, 0.1155097484588623], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [0.07832610607147217, 0.1155097484588623], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001421.pt"}
{"idx": 1423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5iu2FhCiQFw.mp4", "duration": 107.57413333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the girl grab the Santa Claus figure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl grab the Santa Claus figure?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 56.0], "base_se_norm": [-0.014632999897003174, 0.04114246368408203], "gt_se_raw": [98.0, 101.0], "gt_se_norm": [0.8219993114471436, 0.8777748346328735], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001423.pt"}
{"idx": 1425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qcf0KPEM-y8.mp4", "duration": 8.074733333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the boy in the orange shirt looking down at his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the boy in the orange shirt looking down at his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.23843097686767578], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.00925523042678833], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001425.pt"}
{"idx": 1427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qcf0KPEM-y8.mp4", "duration": 8.074733333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time frame where the boy is smiling while looking at the bracelet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time frame where the boy is smiling while looking at the bracelet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.23843097686767578], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.5046275854110718, 0.23843097686767578], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001427.pt"}
{"idx": 1429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99jwJbaMG1Q.mp4", "duration": 69.069, "prompt_text": "Please find the visual event described by the sentence 'When does the dog start chasing the duck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog start chasing the duck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [0.5057406425476074, 0.621566891670227], "gt_se_raw": [52.0, 67.0], "gt_se_norm": [0.5057406425476074, 0.9400888681411743], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001429.pt"}
{"idx": 1431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99jwJbaMG1Q.mp4", "duration": 69.069, "prompt_text": "Please find the visual event described by the sentence 'At what point does the duck first walk into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the duck first walk into the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [0.12930548191070557, 0.21617507934570312], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [0.12930548191070557, 0.1872185468673706], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001431.pt"}
{"idx": 1433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99jwJbaMG1Q.mp4", "duration": 69.069, "prompt_text": "Please find the visual event described by the sentence 'When are all three animals - the dog, the chicken, and the duck - visible together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all three animals - the dog, the chicken, and the duck - visible together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 68.0], "base_se_norm": [0.15826201438903809, 0.9690454006195068], "gt_se_raw": [56.0, 66.0], "gt_se_norm": [0.621566891670227, 0.9111323356628418], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001433.pt"}
{"idx": 1435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LeTwAJ1cRa8.mp4", "duration": 146.71323333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a hand draws a line of diverse people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a hand draws a line of diverse people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [-0.11391764879226685, -0.0184934139251709], "gt_se_raw": [65.0, 69.0], "gt_se_norm": [-0.11391764879226685, -0.05938953161239624], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001435.pt"}
{"idx": 1437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LeTwAJ1cRa8.mp4", "duration": 146.71323333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a woman in a wheelchair boards a bus using a ramp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a woman in a wheelchair boards a bus using a ramp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 122.0], "base_se_norm": [0.5267879962921143, 0.6631083488464355], "gt_se_raw": [118.0, 122.0], "gt_se_norm": [0.6085802316665649, 0.6631083488464355], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001437.pt"}
{"idx": 1439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H5MKL1duTrg.mp4", "duration": 196.59640000000002, "prompt_text": "Please find the visual event described by the sentence 'When is a man in a mask shown completing paperwork at a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man in a mask shown completing paperwork at a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 73.0], "base_se_norm": [-0.28788119554519653, -0.25736182928085327], "gt_se_raw": [70.0, 74.0], "gt_se_norm": [-0.28788119554519653, -0.24718868732452393], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001439.pt"}
{"idx": 1441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H5MKL1duTrg.mp4", "duration": 196.59640000000002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment someone is washing their hands with soap and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment someone is washing their hands with soap and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 107.0], "base_se_norm": [0.02748572826385498, 0.0885244607925415], "gt_se_raw": [101.0, 107.0], "gt_se_norm": [0.02748572826385498, 0.0885244607925415], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001441.pt"}
{"idx": 1443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H5MKL1duTrg.mp4", "duration": 196.59640000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman with purple nail polish using a hand sanitizer pump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman with purple nail polish using a hand sanitizer pump?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 110.0], "base_se_norm": [0.0885244607925415, 0.11904382705688477], "gt_se_raw": [107.0, 110.0], "gt_se_norm": [0.0885244607925415, 0.11904382705688477], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001443.pt"}
{"idx": 1445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O_HyZ5aW76c.mp4", "duration": 94.8948, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man repeatedly asks, 'What does that mean to play us out?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man repeatedly asks, 'What does that mean to play us out?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 58.0], "base_se_norm": [-0.11480921506881714, 0.222406268119812], "gt_se_raw": [25.0, 38.0], "gt_se_norm": [-0.4731007218360901, -0.199113130569458], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001445.pt"}
{"idx": 1447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O_HyZ5aW76c.mp4", "duration": 94.8948, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man exclaims, 'Fucking thing sucks!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man exclaims, 'Fucking thing sucks!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.009429395198822021, 0.053798556327819824], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [0.5174698829650879, 0.5596218109130859], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001447.pt"}
{"idx": 1449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VwJCHnk7C-Q.mp4", "duration": 60.2, "prompt_text": "Please find the visual event described by the sentence 'When does the video introduce the Colosseum?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video introduce the Colosseum?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 14.0], "base_se_norm": [-0.8338870406150818, -0.5348837375640869], "gt_se_raw": [5.0, 15.0], "gt_se_norm": [-0.8338870406150818, -0.5016611218452454], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001449.pt"}
{"idx": 1451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VwJCHnk7C-Q.mp4", "duration": 60.2, "prompt_text": "Please find the visual event described by the sentence 'At what time is the Leaning Tower of Pisa featured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the Leaning Tower of Pisa featured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 34.0], "base_se_norm": [-0.16943520307540894, 0.12956809997558594], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.16943520307540894, 0.16279065608978271], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001451.pt"}
{"idx": 1453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VwJCHnk7C-Q.mp4", "duration": 60.2, "prompt_text": "Please find the visual event described by the sentence 'Locate the part of the video that showcases the Amalfi Coast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part of the video that showcases the Amalfi Coast.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [0.49501657485961914, 0.7940199375152588], "gt_se_raw": [45.0, 55.0], "gt_se_norm": [0.49501657485961914, 0.8272424936294556], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001453.pt"}
{"idx": 1455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L-LE-j2zkCU.mp4", "duration": 159.7012, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman with long brown hair stroking a dog on her lap while sitting at a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman with long brown hair stroking a dog on her lap while sitting at a desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.9373830556869507, -0.8872895240783691], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.6242983937263489, -0.5992516279220581], "start_token_span": [77, 80], "end_token_span": [82, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001455.pt"}
{"idx": 1457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L-LE-j2zkCU.mp4", "duration": 159.7012, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a red shirt standing behind a rope barrier in an apartment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a red shirt standing behind a rope barrier in an apartment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 83.0], "base_se_norm": [-0.11083948612213135, 0.03944110870361328], "gt_se_raw": [78.0, 86.0], "gt_se_norm": [-0.02317577600479126, 0.0770113468170166], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001457.pt"}
{"idx": 1459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iDnE3PV4YNc.mp4", "duration": 193.27641666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the suit start shooting in the parking garage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the suit start shooting in the parking garage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 78.0], "base_se_norm": [-0.275648832321167, -0.192865788936615], "gt_se_raw": [70.0, 79.0], "gt_se_norm": [-0.275648832321167, -0.18251794576644897], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001459.pt"}
{"idx": 1461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iDnE3PV4YNc.mp4", "duration": 193.27641666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of the cassette tape with 'Mom' written on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of the cassette tape with 'Mom' written on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [-0.6585201621055603, -0.6171286702156067], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.6585201621055603, -0.6274765729904175], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001461.pt"}
{"idx": 1463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iDnE3PV4YNc.mp4", "duration": 193.27641666666665, "prompt_text": "Please find the visual event described by the sentence 'When do the man and woman enter the black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the man and woman enter the black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 159.0], "base_se_norm": [0.5935726165771484, 0.6453119516372681], "gt_se_raw": [155.0, 158.0], "gt_se_norm": [0.6039204597473145, 0.634964108467102], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001463.pt"}
{"idx": 1465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TQQXstNh45g.mp4", "duration": 90.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the orangutan examine the shampoo bottle on the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the orangutan examine the shampoo bottle on the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.6888889074325562, -0.6000000238418579], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.6666666269302368, -0.6222221851348877], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001465.pt"}
{"idx": 1467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TQQXstNh45g.mp4", "duration": 90.0, "prompt_text": "Please find the visual event described by the sentence 'When does the girl hold up her drawing that says 'SAVE MY FRIEND'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl hold up her drawing that says 'SAVE MY FRIEND'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 69.0], "base_se_norm": [0.46666669845581055, 0.5333333015441895], "gt_se_raw": [66.0, 68.0], "gt_se_norm": [0.46666669845581055, 0.5111111402511597], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001467.pt"}
{"idx": 1469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TmcuBRa63qg.mp4", "duration": 161.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see Pooh poking a campfire with a stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Pooh poking a campfire with a stick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 47.0], "base_se_norm": [-0.47826087474823, -0.4161490797996521], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.5403726696968079, -0.4658384919166565], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001469.pt"}
{"idx": 1471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TmcuBRa63qg.mp4", "duration": 161.0, "prompt_text": "Please find the visual event described by the sentence 'Show me when Pooh is doing exercises in front of the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when Pooh is doing exercises in front of the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 111.0], "base_se_norm": [-0.31677019596099854, 0.37888193130493164], "gt_se_raw": [77.0, 103.0], "gt_se_norm": [-0.04347825050354004, 0.27950310707092285], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001471.pt"}
{"idx": 1473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TmcuBRa63qg.mp4", "duration": 161.0, "prompt_text": "Please find the visual event described by the sentence 'When does the cuckoo clock with the honey pot appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cuckoo clock with the honey pot appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 51.0], "base_se_norm": [-0.4161490797996521, -0.36645960807800293], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [-0.4161490797996521, -0.3540372848510742], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001473.pt"}
{"idx": 1475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AfW1MDjZB30.mp4", "duration": 47.213833333333326, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a hand in a plaid shirt drops a small object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a hand in a plaid shirt drops a small object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.364592969417572, -0.23751157522201538], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.364592969417572, -0.23751157522201538], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001475.pt"}
{"idx": 1477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AfW1MDjZB30.mp4", "duration": 47.213833333333326, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the text 'SUBSCRIBE TO DAD' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the text 'SUBSCRIBE TO DAD' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 46.0], "base_se_norm": [0.18609309196472168, 0.9485814571380615], "gt_se_raw": [28.0, 47.0], "gt_se_norm": [0.18609309196472168, 0.9909418821334839], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001477.pt"}
{"idx": 1479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0QswzjY6MDU.mp4", "duration": 35.035000000000004, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person introduces all the ingredients for the Chex mix.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person introduces all the ingredients for the Chex mix.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 34.0], "base_se_norm": [-0.14371341466903687, 0.9409161806106567], "gt_se_raw": [14.0, 34.0], "gt_se_norm": [-0.20079922676086426, 0.9409161806106567], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001479.pt"}
{"idx": 1481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0QswzjY6MDU.mp4", "duration": 35.035000000000004, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the bag of Reese's Pieces is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the bag of Reese's Pieces is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.02954190969467163, 0.141715407371521], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.02754390239715576, 0.141715407371521], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001481.pt"}
{"idx": 1483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0QswzjY6MDU.mp4", "duration": 35.035000000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the Honey Nut Chex box is the main focus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the Honey Nut Chex box is the main focus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.7696589231491089, 0.9409161806106567], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [0.7696589231491089, 0.9409161806106567], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001483.pt"}
{"idx": 1485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JK5-2X_GqP8.mp4", "duration": 158.4583, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy stretches the yellow balloon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy stretches the yellow balloon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 62.0], "base_se_norm": [-0.3058110475540161, -0.21745973825454712], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [-0.2931894063949585, -0.2553246021270752], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001485.pt"}
{"idx": 1487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JK5-2X_GqP8.mp4", "duration": 158.4583, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy places the balloon over the bottle's opening.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy places the balloon over the bottle's opening.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 103.0], "base_se_norm": [0.08545911312103271, 0.300026535987854], "gt_se_raw": [85.0, 93.0], "gt_se_norm": [0.0728374719619751, 0.17381036281585693], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001487.pt"}
{"idx": 1489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RBAmLm_jYyY.mp4", "duration": 107.32, "prompt_text": "Please find the visual event described by the sentence 'At what point is a silhouette of a woman walking past a broken fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a silhouette of a woman walking past a broken fence?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.7577338814735413], "gt_se_raw": [2.0, 13.0], "gt_se_norm": [-0.9627282619476318, -0.7577338814735413], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001489.pt"}
{"idx": 1491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RBAmLm_jYyY.mp4", "duration": 107.32, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the woman collapses onto the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the woman collapses onto the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 42.0], "base_se_norm": [-0.49683189392089844, -0.2172940969467163], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.5154677629470825, -0.44092434644699097], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001491.pt"}
{"idx": 1493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RBAmLm_jYyY.mp4", "duration": 107.32, "prompt_text": "Please find the visual event described by the sentence 'When does the woman stand and make her final vow, 'I'll never be hungry again'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman stand and make her final vow, 'I'll never be hungry again'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 83.0], "base_se_norm": [0.45359671115875244, 0.5467760562896729], "gt_se_raw": [77.0, 83.0], "gt_se_norm": [0.43496084213256836, 0.5467760562896729], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001493.pt"}
{"idx": 1495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mc5T4c_mj1U.mp4", "duration": 84.55113333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman is tying her hair up in front of the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman is tying her hair up in front of the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 34.0], "base_se_norm": [-0.40864187479019165, -0.19575291872024536], "gt_se_raw": [26.0, 34.0], "gt_se_norm": [-0.3849875330924988, -0.19575291872024536], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001495.pt"}
{"idx": 1497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mc5T4c_mj1U.mp4", "duration": 84.55113333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the man pulling the woman's hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man pulling the woman's hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [0.34829652309417725, 0.41925954818725586], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [0.34829652309417725, 0.41925954818725586], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001497.pt"}
{"idx": 1499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZqD0iHDKmzg.mp4", "duration": 196.02916666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does Tom add chopped vegetables to the fishbowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Tom add chopped vegetables to the fishbowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 36.0], "base_se_norm": [-0.6837205290794373, -0.6327077150344849], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.7041256427764893, -0.6735179424285889], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001499.pt"}
{"idx": 1501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZqD0iHDKmzg.mp4", "duration": 196.02916666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment Tom's head is looking out of the mouse hole.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment Tom's head is looking out of the mouse hole.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 104.0], "base_se_norm": [-0.020553886890411377, 0.061066627502441406], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.030458927154541016, 0.05086410045623779], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001501.pt"}
{"idx": 1503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZqD0iHDKmzg.mp4", "duration": 196.02916666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where Tom gets flattened by an iron.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 177.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where Tom gets flattened by an iron.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 177.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 177.0], "base_se_norm": [0.7650434970855713, 0.8058538436889648], "gt_se_raw": [175.0, 178.0], "gt_se_norm": [0.7854486703872681, 0.8160563707351685], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001503.pt"}
{"idx": 1505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eSLe4HuKuK0.mp4", "duration": 119.11899999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man lead a herd of reindeer through a forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man lead a herd of reindeer through a forest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 76.0], "base_se_norm": [0.1920851469039917, 0.2760348320007324], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [0.1920851469039917, 0.2592449188232422], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001505.pt"}
{"idx": 1507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eSLe4HuKuK0.mp4", "duration": 119.11899999999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the main characters are looking out over an autumn landscape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the main characters are looking out over an autumn landscape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 96.0], "base_se_norm": [0.4607241153717041, 0.6118334531784058], "gt_se_raw": [87.0, 96.0], "gt_se_norm": [0.4607241153717041, 0.6118334531784058], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001507.pt"}
{"idx": 1509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l-RVtiMtfEs.mp4", "duration": 392.4921, "prompt_text": "Please find the visual event described by the sentence 'When do the two men appear together at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two men appear together at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, -0.8878958225250244], "gt_se_raw": [0.0, 22.0], "gt_se_norm": [-1.0, -0.8878958225250244], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001509.pt"}
{"idx": 1511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l-RVtiMtfEs.mp4", "duration": 392.4921, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man in the red shirt introduces himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man in the red shirt introduces himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 51.0], "base_se_norm": [-0.8012698888778687, -0.7401221990585327], "gt_se_raw": [39.0, 50.0], "gt_se_norm": [-0.8012698888778687, -0.7452178001403809], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001511.pt"}
{"idx": 1513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l-RVtiMtfEs.mp4", "duration": 392.4921, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where an animation of puzzle pieces is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where an animation of puzzle pieces is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 86.0], "base_se_norm": [-0.5872528553009033, -0.5617746114730835], "gt_se_raw": [81.0, 86.0], "gt_se_norm": [-0.5872528553009033, -0.5617746114730835], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001513.pt"}
{"idx": 1515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hEteiEVRBng.mp4", "duration": 251.64, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the man is in an alley and a car screeches to a halt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the man is in an alley and a car screeches to a halt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [-0.5946590304374695, -0.5787633061408997], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [-0.5151804089546204, -0.49928468465805054], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001515.pt"}
{"idx": 1517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hEteiEVRBng.mp4", "duration": 251.64, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the white shirt save the woman from a man with a knife in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the white shirt save the woman from a man with a knife in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 162.0], "base_se_norm": [0.20807504653930664, 0.28755366802215576], "gt_se_raw": [153.0, 157.0], "gt_se_norm": [0.21602284908294678, 0.24781429767608643], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001517.pt"}
{"idx": 1519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2OukFKrEWRs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker notes the significant difference in employment rates between those with and without disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker notes the significant difference in employment rates between those with and without disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9031662940979004, -0.03166300058364868], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9031662940979004, -0.4189978241920471], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001519.pt"}
{"idx": 1521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2OukFKrEWRs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Isolate the moment where the entire narration about the employment data is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Isolate the moment where the entire narration about the employment data is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, 0.355671763420105], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001521.pt"}
{"idx": 1523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mD4mJF7LEJU.mp4", "duration": 30.063366666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the Father's Day card flying through the city.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the Father's Day card flying through the city.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.20168620347976685, 0.06441843509674072], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.5343169569969177, -0.0021077394485473633], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001523.pt"}
{"idx": 1525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mD4mJF7LEJU.mp4", "duration": 30.063366666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the father and son walking in the backyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the father and son walking in the backyard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.26399683952331543, 0.4635753631591797], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [0.26399683952331543, 0.5301014184951782], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001525.pt"}
{"idx": 1527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sIsS73PeaXI.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'When can we see the snowman's arms catch fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the snowman's arms catch fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8167617321014404, -0.7744759917259216], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.8026664853096008, -0.760380744934082], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001527.pt"}
{"idx": 1529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sIsS73PeaXI.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the snowman is juggling ice cream cones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the snowman is juggling ice cream cones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 69.0], "base_se_norm": [-0.15428489446640015, -0.027427613735198975], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.15428489446640015, -0.11199915409088135], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001529.pt"}
{"idx": 1531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sIsS73PeaXI.mp4", "duration": 141.89175, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a man with blond hair is holding two pieces of wood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a man with blond hair is holding two pieces of wood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [-0.3657136559486389, -0.23885637521743774], "gt_se_raw": [45.0, 53.0], "gt_se_norm": [-0.3657136559486389, -0.2529516816139221], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001531.pt"}
{"idx": 1533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epwyvihMOmE.mp4", "duration": 122.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman's face is superimposed on the sun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman's face is superimposed on the sun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 122.0], "base_se_norm": [-1.0, 0.987510085105896], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.5927233695983887, -0.4298126697540283], "start_token_span": [70, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001533.pt"}
{"idx": 1535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epwyvihMOmE.mp4", "duration": 122.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with a bright yellow sun against a blue sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with a bright yellow sun against a blue sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 122.0], "base_se_norm": [-1.0, 0.987510085105896], "gt_se_raw": [80.0, 90.0], "gt_se_norm": [0.3032853603363037, 0.4661959409713745], "start_token_span": [68, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001535.pt"}
{"idx": 1537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EUCsYSC80Xw.mp4", "duration": 98.5985985985986, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman in a blue shirt is speaking in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman in a blue shirt is speaking in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 34.0], "base_se_norm": [-0.9594314694404602, -0.31033504009246826], "gt_se_raw": [2.0, 18.0], "gt_se_norm": [-0.9594314694404602, -0.6348832845687866], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001537.pt"}
{"idx": 1539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EUCsYSC80Xw.mp4", "duration": 98.5985985985986, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman with blonde hair is interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman with blonde hair is interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 83.0], "base_se_norm": [-0.18862944841384888, 0.6835938692092896], "gt_se_raw": [39.0, 46.0], "gt_se_norm": [-0.20891374349594116, -0.06692391633987427], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001539.pt"}
{"idx": 1541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EUCsYSC80Xw.mp4", "duration": 98.5985985985986, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip of a man performing an exercise while lying on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip of a man performing an exercise while lying on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 71.0], "base_se_norm": [0.2981928586959839, 0.4401826858520508], "gt_se_raw": [64.0, 72.0], "gt_se_norm": [0.2981928586959839, 0.46046698093414307], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001541.pt"}
{"idx": 1543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8BbZTCdB2X4.mp4", "duration": 192.1, "prompt_text": "Please find the visual event described by the sentence 'Show me a baby kangaroo in its mother's pouch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me a baby kangaroo in its mother's pouch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 86.0], "base_se_norm": [-0.22956794500350952, -0.10463303327560425], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.583550214767456, -0.5627277493476868], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001543.pt"}
{"idx": 1545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8BbZTCdB2X4.mp4", "duration": 192.1, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a kangaroo leaping over another one?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a kangaroo leaping over another one?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.8021863698959351, -0.7709526419639587], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.8021863698959351, -0.7709526419639587], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001545.pt"}
{"idx": 1547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UbEEdPKBvSU.mp4", "duration": 196.125, "prompt_text": "Please find the visual event described by the sentence 'When is the man walking away from the woman in the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 168.0 - 180.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man walking away from the woman in the room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 168.0 - 180.0 seconds.<|im_end|>\n", "base_se_raw": [168.0, 180.0], "base_se_norm": [0.7131930589675903, 0.8355640172958374], "gt_se_raw": [14.0, 15.0], "gt_se_norm": [-0.8572338819503784, -0.8470363616943359], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001547.pt"}
{"idx": 1549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UbEEdPKBvSU.mp4", "duration": 196.125, "prompt_text": "Please find the visual event described by the sentence 'When does the man hang his coat on the coat rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man hang his coat on the coat rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.9388145208358765, -0.9082217812538147], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.9388145208358765, -0.918419361114502], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001549.pt"}
{"idx": 1551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UbEEdPKBvSU.mp4", "duration": 196.125, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man walk out of the room and into the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man walk out of the room and into the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 183.0], "base_se_norm": [0.8049713373184204, 0.866156816482544], "gt_se_raw": [178.0, 181.0], "gt_se_norm": [0.8151688575744629, 0.8457616567611694], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001551.pt"}
{"idx": 1553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tfIqUsJ_NA.mp4", "duration": 173.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman place her hand on her chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman place her hand on her chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.7225968241691589, -0.6763629913330078], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7225968241691589, -0.687921404838562], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001553.pt"}
{"idx": 1555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tfIqUsJ_NA.mp4", "duration": 173.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man holds his hands together in front of his chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man holds his hands together in front of his chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 104.0], "base_se_norm": [0.1674051284790039, 0.20208048820495605], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.1674051284790039, 0.1905219554901123], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001555.pt"}
{"idx": 1557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GfBVrL6WtjI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what moment is a woman with blonde hair sitting on a stool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is a woman with blonde hair sitting on a stool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001557.pt"}
{"idx": 1559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GfBVrL6WtjI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man wearing glasses is looking thoughtfully at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man wearing glasses is looking thoughtfully at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08188092708587646, 0.3771786689758301], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001559.pt"}
{"idx": 1561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GfBVrL6WtjI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the words 'VISUAL COMMUNICATION' are displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the words 'VISUAL COMMUNICATION' are displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.8688400983810425, -0.5409404635429382], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.8688400983810425, -0.5409404635429382], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001561.pt"}
{"idx": 1563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ye2sScA_bqI.mp4", "duration": 137.37056666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the video reveal that the bus is the main subject?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video reveal that the bus is the main subject?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [-0.4467518925666809, -0.3739560842514038], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.4030744433403015, -0.3739560842514038], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001563.pt"}
{"idx": 1565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ye2sScA_bqI.mp4", "duration": 137.37056666666666, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the operation of the wheelchair lift on the white van is demonstrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the operation of the wheelchair lift on the white van is demonstrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 88.0], "base_se_norm": [-0.009977102279663086, 0.2812061309814453], "gt_se_raw": [68.0, 80.0], "gt_se_norm": [-0.009977102279663086, 0.16473281383514404], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001565.pt"}
{"idx": 1567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgWgJa1NtAY.mp4", "duration": 128.5284, "prompt_text": "Please find the visual event described by the sentence 'When are the boys working together in the garage during the daytime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the boys working together in the garage during the daytime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 68.0], "base_se_norm": [-0.2686440944671631, 0.05813193321228027], "gt_se_raw": [47.0, 58.0], "gt_se_norm": [-0.2686440944671631, -0.09747570753097534], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001567.pt"}
{"idx": 1569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgWgJa1NtAY.mp4", "duration": 128.5284, "prompt_text": "Please find the visual event described by the sentence 'At what point does a boy on a skateboard ride towards the camera and point?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a boy on a skateboard ride towards the camera and point?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 89.0], "base_se_norm": [0.2759826183319092, 0.38490796089172363], "gt_se_raw": [84.0, 89.0], "gt_se_norm": [0.30710411071777344, 0.38490796089172363], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001569.pt"}
{"idx": 1571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgWgJa1NtAY.mp4", "duration": 128.5284, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where three boys are working in the garage at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where three boys are working in the garage at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 42.0], "base_se_norm": [-0.8132708072662354, -0.3464479446411133], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.6265416741371155, -0.5020555257797241], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001571.pt"}
{"idx": 1573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5nmhHL3sVIk.mp4", "duration": 34.7, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl in the driver's seat is actively turning the steering wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl in the driver's seat is actively turning the steering wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 18.0], "base_se_norm": [-0.8270893096923828, 0.03746390342712402], "gt_se_raw": [2.0, 20.0], "gt_se_norm": [-0.8847262263298035, 0.15273773670196533], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001573.pt"}
{"idx": 1575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5nmhHL3sVIk.mp4", "duration": 34.7, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the camera angle changes to a front-facing view of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the camera angle changes to a front-facing view of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 34.0], "base_se_norm": [0.15273773670196533, 0.9596540927886963], "gt_se_raw": [21.0, 34.0], "gt_se_norm": [0.2103745937347412, 0.9596540927886963], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001575.pt"}
{"idx": 1577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Uo0KjdDJr1c.mp4", "duration": 170.83733333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man mention he wants the woman to be ready at 8 sharp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man mention he wants the woman to be ready at 8 sharp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [-0.5668393969535828, -0.5200111865997314], "gt_se_raw": [60.0, 66.0], "gt_se_norm": [-0.2975773811340332, -0.2273351550102234], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001577.pt"}
{"idx": 1579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Uo0KjdDJr1c.mp4", "duration": 170.83733333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man say he doesn't think they'll be a good fit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man say he doesn't think they'll be a good fit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [-0.5668393969535828, -0.46147602796554565], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [0.346310019493103, 0.38143110275268555], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001579.pt"}
{"idx": 1581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Uo0KjdDJr1c.mp4", "duration": 170.83733333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the woman seen making a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman seen making a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 155.0], "base_se_norm": [0.6741071939468384, 0.8145917654037476], "gt_se_raw": [143.0, 150.0], "gt_se_norm": [0.6741071939468384, 0.756056547164917], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001581.pt"}
{"idx": 1583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GyTxtJ4gVLE.mp4", "duration": 110.54387721054388, "prompt_text": "Please find the visual event described by the sentence 'Show the diagram illustrating the process from the brain to the symptoms of Parkinson's.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the diagram illustrating the process from the brain to the symptoms of Parkinson's.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 38.0], "base_se_norm": [-0.4391367435455322, -0.31249016523361206], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.42104434967041016, -0.31249016523361206], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001583.pt"}
{"idx": 1585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GyTxtJ4gVLE.mp4", "duration": 110.54387721054388, "prompt_text": "Please find the visual event described by the sentence 'When is an image of a person struggling to eat soup with a spoon shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is an image of a person struggling to eat soup with a spoon shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.023012399673461914, 0.049357056617736816], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.023012399673461914, 0.049357056617736816], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001585.pt"}
{"idx": 1587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fnd-2jetT1w.mp4", "duration": 259.92633333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the three states of matter, solid, liquid, and gas, shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three states of matter, solid, liquid, and gas, shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.8230267763137817, -0.7845543622970581], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.8230267763137817, -0.7845543622970581], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001587.pt"}
{"idx": 1589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fnd-2jetT1w.mp4", "duration": 259.92633333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the definition of 'viscosity' is explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 207.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the definition of 'viscosity' is explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 207.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 207.0], "base_se_norm": [0.4773416519165039, 0.5927590131759644], "gt_se_raw": [191.0, 195.0], "gt_se_norm": [0.4696471691131592, 0.5004251003265381], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001589.pt"}
{"idx": 1591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fnd-2jetT1w.mp4", "duration": 259.92633333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the word 'INVESTIGATION' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the word 'INVESTIGATION' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 138.0], "base_se_norm": [0.03875589370727539, 0.06183934211730957], "gt_se_raw": [135.0, 138.0], "gt_se_norm": [0.03875589370727539, 0.06183934211730957], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001591.pt"}
{"idx": 1593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PwirCe0230M.mp4", "duration": 33.70033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a woman walks into the frame and stands by the desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a woman walks into the frame and stands by the desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.05045449733734131, 0.3056250810623169], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [0.008892059326171875, 0.18693184852600098], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001593.pt"}
{"idx": 1595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PwirCe0230M.mp4", "duration": 33.70033333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman is operating the laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman is operating the laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 25.0], "base_se_norm": [0.008892059326171875, 0.483664870262146], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.3649716377258301, 0.483664870262146], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001595.pt"}
{"idx": 1597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6d3zmkbef28.mp4", "duration": 39.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the giraffe walking on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the giraffe walking on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 31.0], "base_se_norm": [-1.0, 0.5749365091323853], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.8983911871910095], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001597.pt"}
{"idx": 1599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6d3zmkbef28.mp4", "duration": 39.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the giraffe walk off the road and onto the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the giraffe walk off the road and onto the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [0.4225233793258667, 0.6257410049438477], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.47332775592803955, 0.6257410049438477], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001599.pt"}
{"idx": 1601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6d3zmkbef28.mp4", "duration": 39.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we observe another animal lying down in the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe another animal lying down in the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 39.0], "base_se_norm": [0.8289586305618286, 0.9813717603683472], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [0.6765453815460205, 0.8797630071640015], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001601.pt"}
{"idx": 1603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R9Puz5RFl5o.mp4", "duration": 82.816, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the black shirt appear in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the black shirt appear in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.7585008144378662, -0.6619011163711548], "gt_se_raw": [10.0, 18.0], "gt_se_norm": [-0.7585008144378662, -0.5653014183044434], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001603.pt"}
{"idx": 1605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R9Puz5RFl5o.mp4", "duration": 82.816, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the woman points to the newly drawn straight line.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the woman points to the newly drawn straight line.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [0.23164606094360352, 0.3523956537246704], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [0.23164606094360352, 0.27994585037231445], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001605.pt"}
{"idx": 1607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/llsWcPkQZJ4.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'When does the chicken introduce itself and say it lives in a farmyard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the chicken introduce itself and say it lives in a farmyard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 20.0], "base_se_norm": [-0.8334998488426208, -0.6669996976852417], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.7335997819900513, -0.6669996976852417], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001607.pt"}
{"idx": 1609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/llsWcPkQZJ4.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the video explains the different kinds of food that chickens eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the video explains the different kinds of food that chickens eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 76.0], "base_se_norm": [0.06560099124908447, 0.2654012441635132], "gt_se_raw": [67.0, 76.0], "gt_se_norm": [0.11555111408233643, 0.2654012441635132], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001609.pt"}
{"idx": 1611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/llsWcPkQZJ4.mp4", "duration": 120.12, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the mother chicken sits on the egg to keep it warm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the mother chicken sits on the egg to keep it warm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 116.0], "base_se_norm": [0.8481518030166626, 0.9314018487930298], "gt_se_raw": [110.0, 114.0], "gt_se_norm": [0.831501841545105, 0.898101806640625], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001611.pt"}
{"idx": 1613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nxzo7WLQ50s.mp4", "duration": 13.446766666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the child successfully opens the drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the child successfully opens the drawer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 7.0], "base_se_norm": [-0.2563267946243286, 0.04114246368408203], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.2563267946243286, 0.04114246368408203], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001613.pt"}
{"idx": 1615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nxzo7WLQ50s.mp4", "duration": 13.446766666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the drawer is open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the drawer is open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.10759216547012329, 0.7848156690597534], "gt_se_raw": [5.0, 13.0], "gt_se_norm": [-0.2563267946243286, 0.933550238609314], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001615.pt"}
{"idx": 1617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JMKHbmwltWQ.mp4", "duration": 222.6224, "prompt_text": "Please find the visual event described by the sentence 'When does one woman approach the other and they begin to communicate through touch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does one woman approach the other and they begin to communicate through touch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 13.0], "base_se_norm": [-0.9730485081672668, -0.883210301399231], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.9730485081672668, -0.9281294345855713], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001617.pt"}
{"idx": 1619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JMKHbmwltWQ.mp4", "duration": 222.6224, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the blindfolded person explores a colorful, segmented ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the blindfolded person explores a colorful, segmented ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 48.0], "base_se_norm": [-0.7125177383422852, -0.568776547908783], "gt_se_raw": [33.0, 48.0], "gt_se_norm": [-0.7035338878631592, -0.568776547908783], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001619.pt"}
{"idx": 1621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JMKHbmwltWQ.mp4", "duration": 222.6224, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the two people are folding a piece of yellow paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the two people are folding a piece of yellow paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 142.0], "base_se_norm": [-0.24535900354385376, 0.27570271492004395], "gt_se_raw": [104.0, 118.0], "gt_se_norm": [-0.06568253040313721, 0.06009101867675781], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001621.pt"}
{"idx": 1623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CY0629SKvt8.mp4", "duration": 54.054, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boy starts using both of his hands to play with the beads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boy starts using both of his hands to play with the beads.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.5559995770454407, -0.3339993357658386], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.29699933528900146, -0.1489991545677185], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001623.pt"}
{"idx": 1625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CY0629SKvt8.mp4", "duration": 54.054, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy knock the plate and all the beads off his lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy knock the plate and all the beads off his lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 37.0], "base_se_norm": [0.18400120735168457, 0.3690013885498047], "gt_se_raw": [38.0, 39.0], "gt_se_norm": [0.40600132942199707, 0.443001389503479], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001625.pt"}
{"idx": 1627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFY-vIeN5XA.mp4", "duration": 57.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is a yellow ladder visible next to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a yellow ladder visible next to a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.7220613956451416, -0.513607382774353], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.7220613956451416, -0.5483497381210327], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001627.pt"}
{"idx": 1629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFY-vIeN5XA.mp4", "duration": 57.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a girl with blue painted hands is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a girl with blue painted hands is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.06195712089538574, 0.04226982593536377], "gt_se_raw": [28.0, 29.0], "gt_se_norm": [-0.02721482515335083, 0.007527470588684082], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001629.pt"}
{"idx": 1631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SFY-vIeN5XA.mp4", "duration": 57.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the children stand together in front of their completed artwork?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the children stand together in front of their completed artwork?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [0.7718586921691895, 0.8413434028625488], "gt_se_raw": [52.0, 53.0], "gt_se_norm": [0.8066010475158691, 0.8413434028625488], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001631.pt"}
{"idx": 1633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dv0VfX9U9Fg.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a person's hands are using a braille keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a person's hands are using a braille keyboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.3791518807411194, -0.2098296880722046], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.3791518807411194, -0.2098296880722046], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001633.pt"}
{"idx": 1635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dv0VfX9U9Fg.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a stock market graph is shown on a laptop screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a stock market graph is shown on a laptop screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.128814697265625, 0.29813694953918457], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [0.07237398624420166, 0.29813694953918457], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001635.pt"}
{"idx": 1637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6D4B81EEr30.mp4", "duration": 33.166466666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman guiding the boy's hand with the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman guiding the boy's hand with the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 32.0], "base_se_norm": [-1.0, 0.9296599626541138], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.8190943598747253, -0.6381887197494507], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001637.pt"}
{"idx": 1639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6D4B81EEr30.mp4", "duration": 33.166466666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman helps the boy scoop food from the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman helps the boy scoop food from the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, 0.38694310188293457], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.27637749910354614, -0.09547191858291626], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001639.pt"}
{"idx": 1641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6D4B81EEr30.mp4", "duration": 33.166466666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the boy uses his left hand to touch his cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the boy uses his left hand to touch his cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.38694310188293457, 0.5678486824035645], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.4472450017929077, 0.5678486824035645], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001641.pt"}
{"idx": 1643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DF8_5mGdcq4.mp4", "duration": 79.954875, "prompt_text": "Please find the visual event described by the sentence 'Show the period when the woman's face is visible but her eyes are still closed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period when the woman's face is visible but her eyes are still closed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 57.0], "base_se_norm": [0.17566323280334473, 0.4258042573928833], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [0.2006772756576538, 0.2507054805755615], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001643.pt"}
{"idx": 1645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DF8_5mGdcq4.mp4", "duration": 79.954875, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a synthesized voice says 'Hi, I'm Lazlo'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a synthesized voice says 'Hi, I'm Lazlo'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9749858975410461, -0.7498588562011719], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [0.27571964263916016, 0.32574784755706787], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001645.pt"}
{"idx": 1647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dRaKMNYG3-I.mp4", "duration": 171.62979166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the woman remove her contact lenses and look at her reflection in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman remove her contact lenses and look at her reflection in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.8485111594200134, -0.6737163066864014], "gt_se_raw": [14.0, 36.0], "gt_se_norm": [-0.8368581533432007, -0.580492377281189], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001647.pt"}
{"idx": 1649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dRaKMNYG3-I.mp4", "duration": 171.62979166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman close the bathroom door and open the medicine cabinet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman close the bathroom door and open the medicine cabinet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.33577966690063477, -0.2658616900444031], "gt_se_raw": [56.0, 63.0], "gt_se_norm": [-0.34743261337280273, -0.2658616900444031], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001649.pt"}
{"idx": 1651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dRaKMNYG3-I.mp4", "duration": 171.62979166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the woman fall to the floor and crawl backward in fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman fall to the floor and crawl backward in fear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 133.0], "base_se_norm": [0.42166459560394287, 0.5498474836349487], "gt_se_raw": [120.0, 133.0], "gt_se_norm": [0.3983585834503174, 0.5498474836349487], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001651.pt"}
{"idx": 1653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZtCZGwLH5_o.mp4", "duration": 118.48503333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time period when the boy is clapping his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time period when the boy is clapping his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 48.0], "base_se_norm": [-0.30792945623397827, -0.18977105617523193], "gt_se_raw": [41.0, 48.0], "gt_se_norm": [-0.30792945623397827, -0.18977105617523193], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001653.pt"}
{"idx": 1655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZtCZGwLH5_o.mp4", "duration": 118.48503333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy lie on his back on the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy lie on his back on the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 67.0], "base_se_norm": [-0.12225198745727539, 0.13094449043273926], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [0.14782428741455078, 0.21534335613250732], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001655.pt"}
{"idx": 1657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2I943-gP904.mp4", "duration": 60.685624999999995, "prompt_text": "Please find the visual event described by the sentence 'When does the video suggest putting down the smartphone as a way to manage anxiety?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video suggest putting down the smartphone as a way to manage anxiety?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 16.0], "base_se_norm": [-0.8352163434028625, -0.47269225120544434], "gt_se_raw": [4.0, 16.0], "gt_se_norm": [-0.8681730628013611, -0.47269225120544434], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001657.pt"}
{"idx": 1659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2I943-gP904.mp4", "duration": 60.685624999999995, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a person's hand painting a seascape on a canvas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a person's hand painting a seascape on a canvas?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 36.0], "base_se_norm": [-0.07721143960952759, 0.18644249439239502], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.07721143960952759, 0.08757221698760986], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001659.pt"}
{"idx": 1661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2I943-gP904.mp4", "duration": 60.685624999999995, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range showing a woman in a blue outfit doing yoga in an open field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range showing a woman in a blue outfit doing yoga in an open field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [0.7467069625854492, 0.9114906787872314], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [0.7137502431869507, 0.8785338401794434], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001661.pt"}
{"idx": 1663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFxrHOo3NJ0.mp4", "duration": 224.09053333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the diver entering the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the diver entering the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.7858008742332458, -0.7322510480880737], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.7411760091781616, -0.7233260869979858], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001663.pt"}
{"idx": 1665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFxrHOo3NJ0.mp4", "duration": 224.09053333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a fish bites the food container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a fish bites the food container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 136.0], "base_se_norm": [-0.01825392246246338, 0.21379518508911133], "gt_se_raw": [149.0, 155.0], "gt_se_norm": [0.3298196792602539, 0.38336944580078125], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001665.pt"}
{"idx": 1667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5TiaQeuNO0.mp4", "duration": 178.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man say 'Hello. My name is Inigo Montoya. You killed my father. Prepare to die.' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say 'Hello. My name is Inigo Montoya. You killed my father. Prepare to die.' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 42.0], "base_se_norm": [-0.6189764738082886, -0.5293238162994385], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, -0.8991408348083496], "start_token_span": [84, 88], "end_token_span": [90, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001667.pt"}
{"idx": 1669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5TiaQeuNO0.mp4", "duration": 178.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man with the mustache struggling to get up against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man with the mustache struggling to get up against the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 83.0], "base_se_norm": [-0.6638027429580688, -0.0698542594909668], "gt_se_raw": [62.0, 83.0], "gt_se_norm": [-0.30519235134124756, -0.0698542594909668], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001669.pt"}
{"idx": 1671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5TiaQeuNO0.mp4", "duration": 178.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man with the mustache kills the man in green.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man with the mustache kills the man in green.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 172.0], "base_se_norm": [0.7930519580841064, 0.9275308847427368], "gt_se_raw": [162.0, 169.0], "gt_se_norm": [0.8154650926589966, 0.8939111232757568], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001671.pt"}
{"idx": 1673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WkxE_Fs_mHI.mp4", "duration": 157.07358333333332, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man presents the woman with a bouquet of flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man presents the woman with a bouquet of flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 47.0], "base_se_norm": [-0.503417432308197, -0.4015543460845947], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [-0.4906845688819885, -0.439752995967865], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001673.pt"}
{"idx": 1675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WkxE_Fs_mHI.mp4", "duration": 157.07358333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the man get on one knee and propose with a mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man get on one knee and propose with a mug?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 78.0], "base_se_norm": [-0.09596508741378784, -0.006834864616394043], "gt_se_raw": [71.0, 76.0], "gt_se_norm": [-0.09596508741378784, -0.03230065107345581], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001675.pt"}
{"idx": 1677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vz3zueYyac8.mp4", "duration": 106.70660000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the cartoon character put on a rainbow-colored top hat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cartoon character put on a rainbow-colored top hat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.9250280857086182, -0.8313131332397461], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9250280857086182, -0.8875421285629272], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001677.pt"}
{"idx": 1679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vz3zueYyac8.mp4", "duration": 106.70660000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a whale shark is swimming just below the surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a whale shark is swimming just below the surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 45.0], "base_se_norm": [-0.5126824378967285, -0.15656572580337524], "gt_se_raw": [26.0, 36.0], "gt_se_norm": [-0.5126824378967285, -0.32525259256362915], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001679.pt"}
{"idx": 1681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vz3zueYyac8.mp4", "duration": 106.70660000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the part where small yellow fish are swimming alongside the whale shark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where small yellow fish are swimming alongside the whale shark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 68.0], "base_se_norm": [-0.13782274723052979, 0.27452290058135986], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [0.1808079481124878, 0.27452290058135986], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001681.pt"}
{"idx": 1683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FXkwv_zWL84.mp4", "duration": 197.3972, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the assorted food coloring is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the assorted food coloring is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [-0.5744620561599731, -0.5541983842849731], "gt_se_raw": [43.0, 44.0], "gt_se_norm": [-0.5643302202224731, -0.5541983842849731], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001683.pt"}
{"idx": 1685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FXkwv_zWL84.mp4", "duration": 197.3972, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the man is mixing the liquid and powder with a spatula.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the man is mixing the liquid and powder with a spatula.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 123.0], "base_se_norm": [-0.07800114154815674, 0.24621832370758057], "gt_se_raw": [95.0, 110.0], "gt_se_norm": [-0.03747367858886719, 0.11450409889221191], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001685.pt"}
{"idx": 1687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lrfIVnMGXK0.mp4", "duration": 180.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the three bears trying to lift the couch together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the three bears trying to lift the couch together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 19.0], "base_se_norm": [-0.9777777791023254, -0.7888888716697693], "gt_se_raw": [3.0, 10.0], "gt_se_norm": [-0.9666666388511658, -0.8888888955116272], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001687.pt"}
{"idx": 1689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lrfIVnMGXK0.mp4", "duration": 180.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of the 'Mystique Physique Gym' building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of the 'Mystique Physique Gym' building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.5666666626930237, -0.5222222208976746], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.5666666626930237, -0.5222222208976746], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001689.pt"}
{"idx": 1691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lrfIVnMGXK0.mp4", "duration": 180.0, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the white bear is practicing Tai Chi with a group of elderly people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the white bear is practicing Tai Chi with a group of elderly people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 158.0], "base_se_norm": [0.5888888835906982, 0.7555555105209351], "gt_se_raw": [155.0, 158.0], "gt_se_norm": [0.7222222089767456, 0.7555555105209351], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001691.pt"}
{"idx": 1693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IApvU6SMq-8.mp4", "duration": 217.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man in a wheelchair appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man in a wheelchair appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 30.0], "base_se_norm": [-0.8525798320770264, -0.7235872149467468], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [-0.25368547439575195, -0.22604423761367798], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001693.pt"}
{"idx": 1695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IApvU6SMq-8.mp4", "duration": 217.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When do two men in black robes appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two men in black robes appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 72.0], "base_se_norm": [-0.38267815113067627, -0.3366093039512634], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [-0.37346434593200684, -0.34582310914993286], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001695.pt"}
{"idx": 1697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggIzq8JUK44.mp4", "duration": 185.97745833333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man using real hooves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man using real hooves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.7956741452217102, -0.7419041991233826], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.7849201560020447, -0.7526581883430481], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001697.pt"}
{"idx": 1699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggIzq8JUK44.mp4", "duration": 185.97745833333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man use gloves to simulate a snake sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man use gloves to simulate a snake sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 115.0], "base_se_norm": [0.15067708492279053, 0.23670899868011475], "gt_se_raw": [107.0, 111.0], "gt_se_norm": [0.15067708492279053, 0.19369304180145264], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001699.pt"}
{"idx": 1701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggIzq8JUK44.mp4", "duration": 185.97745833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man walks around in high heels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man walks around in high heels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 161.0], "base_se_norm": [0.6561146974563599, 0.7313926219940186], "gt_se_raw": [154.0, 158.0], "gt_se_norm": [0.6561146974563599, 0.699130654335022], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001701.pt"}
{"idx": 1703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jyhFQVEMhD0.mp4", "duration": 276.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the player in blue get trapped in the hole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the player in blue get trapped in the hole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.6098735928535461, -0.5809752941131592], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.6098735928535461, -0.5809752941131592], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001703.pt"}
{"idx": 1705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jyhFQVEMhD0.mp4", "duration": 276.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the player in red attack their teammate in blue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in red attack their teammate in blue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 168.0], "base_se_norm": [0.11980736255645752, 0.21372663974761963], "gt_se_raw": [162.0, 168.0], "gt_se_norm": [0.17037928104400635, 0.21372663974761963], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001705.pt"}
{"idx": 1707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wn13H7OzGGI.mp4", "duration": 77.91116666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the red shirt walk from left to the center of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the red shirt walk from left to the center of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 8.0], "base_se_norm": [-0.9486594796180725, -0.79463791847229], "gt_se_raw": [2.0, 9.0], "gt_se_norm": [-0.9486594796180725, -0.7689676284790039], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001707.pt"}
{"idx": 1709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wn13H7OzGGI.mp4", "duration": 77.91116666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman picks up a large bone and looks at it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman picks up a large bone and looks at it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 32.0], "base_se_norm": [-0.5892758369445801, -0.17855161428451538], "gt_se_raw": [17.0, 28.0], "gt_se_norm": [-0.563605546951294, -0.28123265504837036], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001709.pt"}
{"idx": 1711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wn13H7OzGGI.mp4", "duration": 77.91116666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman walks off the screen to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman walks off the screen to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 47.0], "base_se_norm": [0.0011402368545532227, 0.20650231838226318], "gt_se_raw": [39.0, 44.0], "gt_se_norm": [0.0011402368545532227, 0.12949156761169434], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001711.pt"}
{"idx": 1713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M0nDEbrp9nM.mp4", "duration": 157.19704, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the little girl is crying while hugging the blue alien.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the little girl is crying while hugging the blue alien.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 76.0], "base_se_norm": [-0.6055905222892761, -0.033060669898986816], "gt_se_raw": [68.0, 76.0], "gt_se_norm": [-0.13484376668930054, -0.033060669898986816], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001713.pt"}
{"idx": 1715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M0nDEbrp9nM.mp4", "duration": 157.19704, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the blue alien wakes up and starts talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the blue alien wakes up and starts talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 128.0], "base_se_norm": [0.3231801986694336, 0.62852942943573], "gt_se_raw": [120.0, 127.0], "gt_se_norm": [0.526746392250061, 0.6158065795898438], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001715.pt"}
{"idx": 1717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nIs4S9YDPRs.mp4", "duration": 147.78096666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man made of sand smashes through the back of the armored truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man made of sand smashes through the back of the armored truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [-0.47219181060791016, -0.4315912127494812], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.5939936637878418, -0.5533930659294128], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001717.pt"}
{"idx": 1719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nIs4S9YDPRs.mp4", "duration": 147.78096666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where Spider-Man is knocked out of the truck and onto the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where Spider-Man is knocked out of the truck and onto the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 117.0], "base_se_norm": [0.17741823196411133, 0.5834245681762695], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.06618553400039673, -0.039118409156799316], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001719.pt"}
{"idx": 1721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nIs4S9YDPRs.mp4", "duration": 147.78096666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when Spider-Man is sitting on a rooftop, taking off his mask and pouring sand out of his boot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when Spider-Man is sitting on a rooftop, taking off his mask and pouring sand out of his boot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 146.0], "base_se_norm": [0.7999612092971802, 0.9758973121643066], "gt_se_raw": [135.0, 146.0], "gt_se_norm": [0.8270282745361328, 0.9758973121643066], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001721.pt"}
{"idx": 1723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y4baNTo5q4A.mp4", "duration": 62.020291666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman's expression change from worried to mischievous as she makes a decision?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman's expression change from worried to mischievous as she makes a decision?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.12931722402572632, -0.03257465362548828], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.12931722402572632, -0.03257465362548828], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001723.pt"}
{"idx": 1725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y4baNTo5q4A.mp4", "duration": 62.020291666666665, "prompt_text": "Please find the visual event described by the sentence 'When is the box of wine placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the box of wine placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [0.06416785717010498, 0.19315791130065918], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.06416785717010498, 0.12866294384002686], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001725.pt"}
{"idx": 1727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ppe_c4Ij0g8.mp4", "duration": 154.45445445445446, "prompt_text": "Please find the visual event described by the sentence 'When can we see a group photo of soldiers in camouflage uniforms posing against a brick wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a group photo of soldiers in camouflage uniforms posing against a brick wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.8964095711708069, -0.831665575504303], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.8964095711708069, -0.8446143865585327], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001727.pt"}
{"idx": 1729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ppe_c4Ij0g8.mp4", "duration": 154.45445445445446, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a man and a child are practicing baseball on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a man and a child are practicing baseball on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 82.0], "base_se_norm": [0.03590404987335205, 0.06180167198181152], "gt_se_raw": [80.0, 82.0], "gt_se_norm": [0.03590404987335205, 0.06180167198181152], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001729.pt"}
{"idx": 1731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ppe_c4Ij0g8.mp4", "duration": 154.45445445445446, "prompt_text": "Please find the visual event described by the sentence 'When are several young women gathered around and petting the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are several young women gathered around and petting the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 119.0], "base_se_norm": [0.476163387298584, 0.5409073829650879], "gt_se_raw": [114.0, 119.0], "gt_se_norm": [0.476163387298584, 0.5409073829650879], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001731.pt"}
{"idx": 1733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RAqA6gtdalE.mp4", "duration": 150.96, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a person's profile is silhouetted against a cool, blue-lit background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a person's profile is silhouetted against a cool, blue-lit background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 150.0], "base_se_norm": [-0.9867514371871948, 0.987281322479248], "gt_se_raw": [58.0, 63.0], "gt_se_norm": [-0.2315845489501953, -0.16534185409545898], "start_token_span": [76, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001733.pt"}
{"idx": 1735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RAqA6gtdalE.mp4", "duration": 150.96, "prompt_text": "Please find the visual event described by the sentence 'At what point is the character shown in extreme low light, with only their outline visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the character shown in extreme low light, with only their outline visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 150.0], "base_se_norm": [-1.0, 0.987281322479248], "gt_se_raw": [100.0, 104.0], "gt_se_norm": [0.3248542547225952, 0.37784838676452637], "start_token_span": [79, 82], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001735.pt"}
{"idx": 1737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vv2qnoUfjPU.mp4", "duration": 142.39225, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a red dress crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a red dress crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [-0.3819888234138489, -0.3398516774177551], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.3679431080818176, -0.3398516774177551], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001737.pt"}
{"idx": 1739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vv2qnoUfjPU.mp4", "duration": 142.39225, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man with a red beanie is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man with a red beanie is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 75.0], "base_se_norm": [0.025336742401123047, 0.05342817306518555], "gt_se_raw": [73.0, 74.0], "gt_se_norm": [0.025336742401123047, 0.0393824577331543], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001739.pt"}
{"idx": 1741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vv2qnoUfjPU.mp4", "duration": 142.39225, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a man and a woman are hugging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a man and a woman are hugging.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 138.0], "base_se_norm": [0.8540335893630981, 0.9383078813552856], "gt_se_raw": [133.0, 138.0], "gt_se_norm": [0.8680793046951294, 0.9383078813552856], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001741.pt"}
{"idx": 1743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EpcDZbXslfw.mp4", "duration": 180.71031812826124, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man is looking for the snake with a stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man is looking for the snake with a stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 50.0], "base_se_norm": [-0.6347745656967163, -0.4466281533241272], "gt_se_raw": [41.0, 49.0], "gt_se_norm": [-0.5462350845336914, -0.4576956033706665], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001743.pt"}
{"idx": 1745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EpcDZbXslfw.mp4", "duration": 180.71031812826124, "prompt_text": "Please find the visual event described by the sentence 'At what point is the patient being wheeled down a hospital corridor on a gurney?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the patient being wheeled down a hospital corridor on a gurney?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 129.0], "base_se_norm": [0.3723621368408203, 0.42769932746887207], "gt_se_raw": [124.0, 129.0], "gt_se_norm": [0.3723621368408203, 0.42769932746887207], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001745.pt"}
{"idx": 1747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NvgYhf2LnVI.mp4", "duration": 145.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a person standing on a hill with a torch, with the forest on fire in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a person standing on a hill with a torch, with the forest on fire in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 117.0], "base_se_norm": [0.5778641700744629, 0.6053053140640259], "gt_se_raw": [115.0, 117.0], "gt_se_norm": [0.5778641700744629, 0.6053053140640259], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001747.pt"}
{"idx": 1749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NvgYhf2LnVI.mp4", "duration": 145.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person draws and aims a bow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person draws and aims a bow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.3002515435218811, -0.2316487431526184], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.2590898275375366, -0.2316487431526184], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001749.pt"}
{"idx": 1751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NvgYhf2LnVI.mp4", "duration": 145.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the character with an eyepatch press a button and then point at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character with an eyepatch press a button and then point at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 123.0], "base_se_norm": [0.6601876020431519, 0.6876286268234253], "gt_se_raw": [121.0, 122.0], "gt_se_norm": [0.6601876020431519, 0.6739081144332886], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001751.pt"}
{"idx": 1753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rvsuYssoTw0.mp4", "duration": 158.95045833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the green top fall onto the conveyor belt in the supermarket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the green top fall onto the conveyor belt in the supermarket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.28279536962509155, -0.21988272666931152], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.2702128291130066, -0.2324652075767517], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001753.pt"}
{"idx": 1755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rvsuYssoTw0.mp4", "duration": 158.95045833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman in the bar pulls out and shows a red card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman in the bar pulls out and shows a red card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 90.0], "base_se_norm": [0.08209824562072754, 0.13242828845977783], "gt_se_raw": [87.0, 89.0], "gt_se_norm": [0.0946807861328125, 0.11984574794769287], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001755.pt"}
{"idx": 1757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KK-Mff60ZIE.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a variety of custom-made controllers on a website?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a variety of custom-made controllers on a website?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 86.0], "base_se_norm": [0.060773491859436035, 0.14033150672912598], "gt_se_raw": [80.0, 86.0], "gt_se_norm": [0.060773491859436035, 0.14033150672912598], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001757.pt"}
{"idx": 1759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KK-Mff60ZIE.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the pie chart illustrating the percentage of disabled gamers displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the pie chart illustrating the percentage of disabled gamers displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 57.0], "base_se_norm": [-0.36353588104248047, -0.24419885873794556], "gt_se_raw": [49.0, 57.0], "gt_se_norm": [-0.35027623176574707, -0.24419885873794556], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001759.pt"}
{"idx": 1761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KK-Mff60ZIE.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when cables are being plugged into the multiple ports of the adaptive controller.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when cables are being plugged into the multiple ports of the adaptive controller.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 122.0], "base_se_norm": [0.5248619318008423, 0.6176795959472656], "gt_se_raw": [113.0, 120.0], "gt_se_norm": [0.4983426332473755, 0.5911602973937988], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001761.pt"}
{"idx": 1763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9newNjIco3Y.mp4", "duration": 121.6, "prompt_text": "Please find the visual event described by the sentence 'When does the man touch the boy's chin with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man touch the boy's chin with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.7697368264198303, -0.7203947305679321], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [-0.7532894611358643, -0.7368420958518982], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001763.pt"}
{"idx": 1765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9newNjIco3Y.mp4", "duration": 121.6, "prompt_text": "Please find the visual event described by the sentence 'Show me when the tablet displays a colorful, moving image.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the tablet displays a colorful, moving image.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.5723683834075928, -0.5065789222717285], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [0.11842107772827148, 0.18421053886413574], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001765.pt"}
{"idx": 1767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dx0-pNkwOv8.mp4", "duration": 139.80633333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the speaker introducing the topic of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the speaker introducing the topic of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9427779912948608], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9427779912948608], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001767.pt"}
{"idx": 1769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dx0-pNkwOv8.mp4", "duration": 139.80633333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the animated teacher point to the words 'The Disability Discrimination Act' on the chalkboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the animated teacher point to the words 'The Disability Discrimination Act' on the chalkboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.1845862865447998, -0.11305880546569824], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.1845862865447998, -0.09875327348709106], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001769.pt"}
{"idx": 1771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dx0-pNkwOv8.mp4", "duration": 139.80633333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the steps for what to do after experiencing discrimination are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the steps for what to do after experiencing discrimination are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 127.0], "base_se_norm": [0.3876338005065918, 0.8167989253997803], "gt_se_raw": [96.0, 127.0], "gt_se_norm": [0.3733283281326294, 0.8167989253997803], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001771.pt"}
{"idx": 1773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lGcSN2H8MNk.mp4", "duration": 27.3273, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a man in a red T-shirt is holding up a measuring cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a man in a red T-shirt is holding up a measuring cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 16.0], "base_se_norm": [0.09780335426330566, 0.1709902286529541], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [0.09780335426330566, 0.24417710304260254], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001773.pt"}
{"idx": 1775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lGcSN2H8MNk.mp4", "duration": 27.3273, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a person is slicing an onion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a person is slicing an onion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.3413180112838745, -0.19494420289993286], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.2681311368942261, -0.12175732851028442], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001775.pt"}
{"idx": 1777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y07at1bU89Q.mp4", "duration": 155.04, "prompt_text": "Please find the visual event described by the sentence 'Show the man with the beard looking through binoculars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man with the beard looking through binoculars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.6646026372909546, -0.6388028860092163], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.6646026372909546, -0.6388028860092163], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001777.pt"}
{"idx": 1779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y07at1bU89Q.mp4", "duration": 155.04, "prompt_text": "Please find the visual event described by the sentence 'At what point do the men walk past the Indian restaurant set?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the men walk past the Indian restaurant set?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 84.0], "base_se_norm": [0.01909184455871582, 0.08359134197235107], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.01909184455871582, 0.07069146633148193], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001779.pt"}
{"idx": 1781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y07at1bU89Q.mp4", "duration": 155.04, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the men are herded into the pub enclosure and start toasting with drinks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the men are herded into the pub enclosure and start toasting with drinks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 148.0], "base_se_norm": [0.7285861968994141, 0.9091848134994507], "gt_se_raw": [132.0, 147.0], "gt_se_norm": [0.7027864456176758, 0.8962849378585815], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001781.pt"}
{"idx": 1783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/agqgBkpbCoY.mp4", "duration": 164.8647, "prompt_text": "Please find the visual event described by the sentence 'When can we see a diver swimming alongside a whale shark?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a diver swimming alongside a whale shark?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 67.0], "base_se_norm": [-0.2842615842819214, -0.1872122883796692], "gt_se_raw": [59.0, 68.0], "gt_se_norm": [-0.2842615842819214, -0.17508113384246826], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001783.pt"}
{"idx": 1785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/agqgBkpbCoY.mp4", "duration": 164.8647, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time a hammerhead shark is shown swimming in an aquarium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time a hammerhead shark is shown swimming in an aquarium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 93.0], "base_se_norm": [0.031148552894592285, 0.1281977891921997], "gt_se_raw": [85.0, 93.0], "gt_se_norm": [0.031148552894592285, 0.1281977891921997], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001785.pt"}
{"idx": 1787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YFB1moyD1UY.mp4", "duration": 113.6, "prompt_text": "Please find the visual event described by the sentence 'When does the woman introduce herself and state her job titles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman introduce herself and state her job titles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9823943376541138, -0.8239436745643616], "gt_se_raw": [1.0, 7.0], "gt_se_norm": [-0.9823943376541138, -0.8767605423927307], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001787.pt"}
{"idx": 1789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YFB1moyD1UY.mp4", "duration": 113.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the speaker lists examples of disabilities, starting with anxiety and depression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the speaker lists examples of disabilities, starting with anxiety and depression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 40.0], "base_se_norm": [-0.5246478915214539, -0.2957746386528015], "gt_se_raw": [38.0, 47.0], "gt_se_norm": [-0.3309859037399292, -0.17253518104553223], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001789.pt"}
{"idx": 1791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YFB1moyD1UY.mp4", "duration": 113.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the contact information is displayed on a white screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the contact information is displayed on a white screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 113.0], "base_se_norm": [0.8309859037399292, 0.9894366264343262], "gt_se_raw": [103.0, 107.0], "gt_se_norm": [0.8133803606033325, 0.8838028907775879], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001791.pt"}
{"idx": 1793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcNcX0hEr_s.mp4", "duration": 88.1, "prompt_text": "Please find the visual event described by the sentence 'When is the left corner folded up to meet the top point?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the left corner folded up to meet the top point?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 25.0], "base_se_norm": [-0.7048808336257935, -0.4324631094932556], "gt_se_raw": [25.0, 34.0], "gt_se_norm": [-0.4324631094932556, -0.22814983129501343], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001793.pt"}
{"idx": 1795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcNcX0hEr_s.mp4", "duration": 88.1, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the second top flap is folded down to secure the cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the second top flap is folded down to secure the cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 73.0], "base_se_norm": [0.4301929473876953, 0.657207727432251], "gt_se_raw": [61.0, 67.0], "gt_se_norm": [0.3847900629043579, 0.5209988355636597], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001795.pt"}
{"idx": 1797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8jPbRnBwmXI.mp4", "duration": 67.93453333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we first see the woman in the purple shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we first see the woman in the purple shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 46.0], "base_se_norm": [-0.9705598950386047, 0.3542449474334717], "gt_se_raw": [0.0, 46.0], "gt_se_norm": [-1.0, 0.3542449474334717], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001797.pt"}
{"idx": 1799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8jPbRnBwmXI.mp4", "duration": 67.93453333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in purple return to the screen after the photo is shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in purple return to the screen after the photo is shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 67.0], "base_se_norm": [0.8252867460250854, 0.9724873304367065], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.8252867460250854, 0.9724873304367065], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001799.pt"}
{"idx": 1801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8jPbRnBwmXI.mp4", "duration": 67.93453333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker says the family didn't fight with the girl about her behavior.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker says the family didn't fight with the girl about her behavior.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.11679673194885254, 0.0009636878967285156], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [0.7664065361022949, 0.8252867460250854], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001801.pt"}
{"idx": 1803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O7_TzpkCZrg.mp4", "duration": 45.07836666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention the first experience of sitting on the bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention the first experience of sitting on the bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9556328058242798, -0.5563281774520874], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.8668984770774841, -0.7337969541549683], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001803.pt"}
{"idx": 1805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O7_TzpkCZrg.mp4", "duration": 45.07836666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the speaker describes exploring the ride's size and location.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the speaker describes exploring the ride's size and location.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.46759384870529175, -0.02392202615737915], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [0.15354669094085693, 0.3310154676437378], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001805.pt"}
{"idx": 1807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VD0zOrDEvAc.mp4", "duration": 189.05553333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man introduce himself as Sonny Singh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man introduce himself as Sonny Singh?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9576843976974487], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9894210696220398, -0.9788421988487244], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001807.pt"}
{"idx": 1809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VD0zOrDEvAc.mp4", "duration": 189.05553333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the definition of 'Cisgender' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the definition of 'Cisgender' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.6826329231262207, -0.6508961915969849], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.6826329231262207, -0.6508961915969849], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001809.pt"}
{"idx": 1811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VD0zOrDEvAc.mp4", "duration": 189.05553333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker discusses his cisgender maleness and heterosexuality as privileges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker discusses his cisgender maleness and heterosexuality as privileges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 53.0], "base_se_norm": [-0.6403173208236694, -0.4393181800842285], "gt_se_raw": [156.0, 159.0], "gt_se_norm": [0.6503088474273682, 0.6820454597473145], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001811.pt"}
{"idx": 1813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VZaqHyHFCzc.mp4", "duration": 150.76, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where Sandman attacks the security guards inside the truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where Sandman attacks the security guards inside the truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 27.0], "base_se_norm": [-0.7877420783042908, -0.6418148279190063], "gt_se_raw": [17.0, 26.0], "gt_se_norm": [-0.7744759917259216, -0.6550809144973755], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001813.pt"}
{"idx": 1815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VZaqHyHFCzc.mp4", "duration": 150.76, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Spider-Man slides on the street using a piece of a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Spider-Man slides on the street using a piece of a car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 64.0], "base_se_norm": [-0.2172989845275879, -0.15096837282180786], "gt_se_raw": [64.0, 68.0], "gt_se_norm": [-0.15096837282180786, -0.09790390729904175], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001815.pt"}
{"idx": 1817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i0Oj9HbsSCo.mp4", "duration": 9.609, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the girl is brushing her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the girl is brushing her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.8732438087463379], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.6651055812835693], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001817.pt"}
{"idx": 1819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i0Oj9HbsSCo.mp4", "duration": 9.609, "prompt_text": "Please find the visual event described by the sentence 'At what point is the girl attempting to brush the left side of her hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the girl attempting to brush the left side of her hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, 0.6651055812835693], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.5837236046791077, 0.04069101810455322], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001819.pt"}
{"idx": 1821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w_y8z7jiBmk.mp4", "duration": 192.65913333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the chef starts seasoning the raw chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the chef starts seasoning the raw chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 65.0], "base_se_norm": [-0.4913296103477478, -0.3252331614494324], "gt_se_raw": [47.0, 54.0], "gt_se_norm": [-0.5120916366577148, -0.43942445516586304], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001821.pt"}
{"idx": 1823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w_y8z7jiBmk.mp4", "duration": 192.65913333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the chicken is dipped into the egg wash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the chicken is dipped into the egg wash.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 141.0], "base_se_norm": [0.40143883228302, 0.46372509002685547], "gt_se_raw": [136.0, 142.0], "gt_se_norm": [0.4118199348449707, 0.4741060733795166], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001823.pt"}
{"idx": 1825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pGQd630EpLU.mp4", "duration": 13.833333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the person organizing the stamps in the book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person organizing the stamps in the book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.3012048006057739], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.5662650465965271, 0.15662658214569092], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001825.pt"}
{"idx": 1827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pGQd630EpLU.mp4", "duration": 13.833333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the person look at the stamps on the page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person look at the stamps on the page?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.3012048006057739], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.5662650465965271], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001827.pt"}
{"idx": 1829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pGQd630EpLU.mp4", "duration": 13.833333333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the person places a stamp in the middle of the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the person places a stamp in the middle of the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.42168670892715454, -0.1325300931930542], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.27710843086242676, -0.1325300931930542], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001829.pt"}
{"idx": 1831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FOjYigV2FhA.mp4", "duration": 165.0649, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention the G7 or the Nuke Group for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention the G7 or the Nuke Group for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [-0.757671058177948, -0.6849724054336548], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.8061368465423584, -0.7819039821624756], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001831.pt"}
{"idx": 1833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FOjYigV2FhA.mp4", "duration": 165.0649, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the speaker claims the groups are not 'really legit'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the speaker claims the groups are not 'really legit'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 115.0], "base_se_norm": [0.32069265842437744, 0.39339137077331543], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.382061243057251, -0.3578283190727234], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001833.pt"}
{"idx": 1835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KVu2pLGGSNg.mp4", "duration": 175.375, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man close his eyes and smile while looking up at the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man close his eyes and smile while looking up at the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.8517462611198425, -0.7719173431396484], "gt_se_raw": [13.0, 20.0], "gt_se_norm": [-0.8517462611198425, -0.7719173431396484], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001835.pt"}
{"idx": 1837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KVu2pLGGSNg.mp4", "duration": 175.375, "prompt_text": "Please find the visual event described by the sentence 'When can we see the zombie banging on the rear window of the van?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the zombie banging on the rear window of the van?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 62.0], "base_se_norm": [-0.4868139624595642, -0.29294371604919434], "gt_se_raw": [45.0, 53.0], "gt_se_norm": [-0.4868139624595642, -0.39558088779449463], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001837.pt"}
{"idx": 1839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KVu2pLGGSNg.mp4", "duration": 175.375, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man drinks from the creek and then spits the water out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man drinks from the creek and then spits the water out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 158.0], "base_se_norm": [0.5509622097015381, 0.8018531799316406], "gt_se_raw": [140.0, 147.0], "gt_se_norm": [0.5965787172317505, 0.6764076948165894], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001839.pt"}
{"idx": 1841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zVXT9iBoejQ.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator state that the employment rate for non-disabled people is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator state that the employment rate for non-disabled people is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9030095934867859, -0.4180576801300049], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9030095934867859, -0.4180576801300049], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001841.pt"}
{"idx": 1843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zVXT9iBoejQ.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where it is mentioned that people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where it is mentioned that people with disabilities are employed at 'nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.030096113681793213, 0.2608751058578491], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.030096113681793213, 0.35786545276641846], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001843.pt"}
{"idx": 1845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iB52iP2a_MY.mp4", "duration": 95.12836666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the parrot first get a close look at the stuffed bunny in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the parrot first get a close look at the stuffed bunny in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 18.0], "base_se_norm": [-0.8528304100036621, -0.6215639710426331], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.8528304100036621, -0.768733561038971], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001845.pt"}
{"idx": 1847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iB52iP2a_MY.mp4", "duration": 95.12836666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the parrot pecks the stuffed bunny while it's being held up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the parrot pecks the stuffed bunny while it's being held up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [0.5557925701141357, 0.6398894786834717], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [0.030187010765075684, 0.0722353458404541], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001847.pt"}
{"idx": 1849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iB52iP2a_MY.mp4", "duration": 95.12836666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the stuffed bunny placed on the counter in front of the parrot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the stuffed bunny placed on the counter in front of the parrot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 83.0], "base_se_norm": [0.5557925701141357, 0.745010495185852], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.6609137058258057, 0.745010495185852], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001849.pt"}
{"idx": 1851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X9nPJamttwM.mp4", "duration": 171.8717, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a presentation slide explains the features of accessible buses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a presentation slide explains the features of accessible buses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 103.0], "base_se_norm": [-0.057436466217041016, 0.1985684633255005], "gt_se_raw": [81.0, 103.0], "gt_se_norm": [-0.057436466217041016, 0.1985684633255005], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001851.pt"}
{"idx": 1853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X9nPJamttwM.mp4", "duration": 171.8717, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that shows the inaccessible stairs at Redfern station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that shows the inaccessible stairs at Redfern station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 156.0], "base_se_norm": [0.757124662399292, 0.8153074979782104], "gt_se_raw": [150.0, 156.0], "gt_se_norm": [0.7454880475997925, 0.8153074979782104], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001853.pt"}
{"idx": 1855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xBD_s0RhUko.mp4", "duration": 180.63879166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman apply makeup to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman apply makeup to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 15.0], "base_se_norm": [-0.9778563380241394, -0.8339227437973022], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.9778563380241394, -0.933569073677063], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001855.pt"}
{"idx": 1857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xBD_s0RhUko.mp4", "duration": 180.63879166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man stands alone in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man stands alone in the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 79.0], "base_se_norm": [-0.6235581636428833, -0.12532633543014526], "gt_se_raw": [34.0, 80.0], "gt_se_norm": [-0.6235581636428833, -0.11425447463989258], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001857.pt"}
{"idx": 1859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xBD_s0RhUko.mp4", "duration": 180.63879166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the blue polka dot dress hugs the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the blue polka dot dress hugs the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 127.0], "base_se_norm": [0.3729054927825928, 0.40612101554870605], "gt_se_raw": [124.0, 127.0], "gt_se_norm": [0.3729054927825928, 0.40612101554870605], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001859.pt"}
{"idx": 1861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uZ8lcB2IL6Q.mp4", "duration": 133.68, "prompt_text": "Please find the visual event described by the sentence 'When is the yellow wheelchair-accessible swing shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the yellow wheelchair-accessible swing shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.5212447643280029, -0.4763614535331726], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.5212447643280029, -0.4763614535331726], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001861.pt"}
{"idx": 1863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uZ8lcB2IL6Q.mp4", "duration": 133.68, "prompt_text": "Please find the visual event described by the sentence 'Find the shot of the accessible toilet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot of the accessible toilet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 75.0], "base_se_norm": [0.017354965209960938, 0.12208259105682373], "gt_se_raw": [68.0, 75.0], "gt_se_norm": [0.017354965209960938, 0.12208259105682373], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001863.pt"}
{"idx": 1865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yHsRjoRec8.mp4", "duration": 85.3853, "prompt_text": "Please find the visual event described by the sentence 'When does the boy place the number 12 tile on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy place the number 12 tile on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.8126140832901001, -0.7189211845397949], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.7423443794250488, -0.6954979300498962], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001865.pt"}
{"idx": 1867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yHsRjoRec8.mp4", "duration": 85.3853, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where the woman asks what comes before 12.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where the woman asks what comes before 12.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 34.0], "base_se_norm": [-0.3441494107246399, -0.20361000299453735], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.03964734077453613, 0.0071991682052612305], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001867.pt"}
{"idx": 1869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yHsRjoRec8.mp4", "duration": 85.3853, "prompt_text": "Please find the visual event described by the sentence 'When do the woman and the boy high-five to celebrate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the woman and the boy high-five to celebrate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 81.0], "base_se_norm": [0.8504356145858765, 0.8972821235656738], "gt_se_raw": [79.0, 81.0], "gt_se_norm": [0.8504356145858765, 0.8972821235656738], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001869.pt"}
{"idx": 1871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lozRy6VCLCI.mp4", "duration": 120.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'At what time is the man with the white beard and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the man with the white beard and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 25.0], "base_se_norm": [-0.6503496170043945, -0.5837495923042297], "gt_se_raw": [18.0, 26.0], "gt_se_norm": [-0.7002997398376465, -0.5670995712280273], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001871.pt"}
{"idx": 1873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lozRy6VCLCI.mp4", "duration": 120.11999999999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where there's a close-up of the music stand with 'UWF MUSIC' written on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where there's a close-up of the music stand with 'UWF MUSIC' written on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 42.0], "base_se_norm": [-0.3339993357658386, -0.30069929361343384], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.31734931468963623, -0.28404927253723145], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001873.pt"}
{"idx": 1875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zYX7iexkODw.mp4", "duration": 119.88643333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the father try to pull a sweater over the child's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the father try to pull a sweater over the child's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9833175539970398, -0.9165877103805542], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9833175539970398, -0.9332702159881592], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001875.pt"}
{"idx": 1877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zYX7iexkODw.mp4", "duration": 119.88643333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the father successfully put the shoe on the child's right foot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the father successfully put the shoe on the child's right foot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 76.0], "base_se_norm": [0.0009472370147705078, 0.26786649227142334], "gt_se_raw": [60.0, 69.0], "gt_se_norm": [0.0009472370147705078, 0.15108931064605713], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001877.pt"}
{"idx": 1879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zYX7iexkODw.mp4", "duration": 119.88643333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the older boy holds up the picture board and points to it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the older boy holds up the picture board and points to it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 117.0], "base_se_norm": [0.8517524003982544, 0.9518471956253052], "gt_se_raw": [112.0, 118.0], "gt_se_norm": [0.8684349060058594, 0.9685295820236206], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001879.pt"}
{"idx": 1881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MqIJKnUkGLY.mp4", "duration": 237.94604166666664, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the man and a young girl are on the swings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the man and a young girl are on the swings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.8066788911819458, -0.7898683547973633], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.8066788911819458, -0.7898683547973633], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001881.pt"}
{"idx": 1883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MqIJKnUkGLY.mp4", "duration": 237.94604166666664, "prompt_text": "Please find the visual event described by the sentence 'When does the lawyer start asking the man a series of questions that he can't answer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the lawyer start asking the man a series of questions that he can't answer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 106.0], "base_se_norm": [-0.46206289529800415, -0.10904169082641602], "gt_se_raw": [125.0, 130.0], "gt_se_norm": [0.050658345222473145, 0.09268474578857422], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001883.pt"}
{"idx": 1885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z5UScMQUO6Q.mp4", "duration": 105.77233333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point is a young boy shown eating his meal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a young boy shown eating his meal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 35.0], "base_se_norm": [-0.4327439069747925, -0.33820122480392456], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [-0.4327439069747925, -0.33820122480392456], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001885.pt"}
{"idx": 1887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z5UScMQUO6Q.mp4", "duration": 105.77233333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the colorful vegetables being stir-fried in a pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the colorful vegetables being stir-fried in a pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [0.07778656482696533, 0.17232930660247803], "gt_se_raw": [56.0, 61.0], "gt_se_norm": [0.05887806415557861, 0.15342068672180176], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001887.pt"}
{"idx": 1889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z5UScMQUO6Q.mp4", "duration": 105.77233333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene that shows examples of foods to avoid, specifically french fries and chocolates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene that shows examples of foods to avoid, specifically french fries and chocolates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 97.0], "base_se_norm": [0.6828597784042358, 0.8341280221939087], "gt_se_raw": [89.0, 97.0], "gt_se_norm": [0.6828597784042358, 0.8341280221939087], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001889.pt"}
{"idx": 1891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdD8dLkEOAU.mp4", "duration": 126.70719520319788, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl pours a large amount of water onto the unicorn toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl pours a large amount of water onto the unicorn toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.5896049737930298, -0.5264673233032227], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.5738205909729004, -0.542251706123352], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001891.pt"}
{"idx": 1893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdD8dLkEOAU.mp4", "duration": 126.70719520319788, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the girl sits on the fat unicorn's back, trying to make it fly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the girl sits on the fat unicorn's back, trying to make it fly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [-0.16342556476593018, -0.0845034122467041], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [-0.147641122341156, -0.0845034122467041], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001893.pt"}
{"idx": 1895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfvyfN8ycvw.mp4", "duration": 61.93687499999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see a pepperoni pizza on a pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a pepperoni pizza on a pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9677090644836426, -0.7416724562644958], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9677090644836426, -0.7416724562644958], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001895.pt"}
{"idx": 1897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfvyfN8ycvw.mp4", "duration": 61.93687499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does a person gently touch a star-shaped flower in a garden?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a person gently touch a star-shaped flower in a garden?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [-0.06356269121170044, 0.13018298149108887], "gt_se_raw": [28.0, 35.0], "gt_se_norm": [-0.09585362672805786, 0.13018298149108887], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001897.pt"}
{"idx": 1899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfvyfN8ycvw.mp4", "duration": 61.93687499999999, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a person is using a black pen-like device on a tactile board with white squares.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a person is using a black pen-like device on a tactile board with white squares.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [0.7437108755111694, 0.937456488609314], "gt_se_raw": [54.0, 60.0], "gt_se_norm": [0.7437108755111694, 0.937456488609314], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001899.pt"}
{"idx": 1901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3BCIO1uX6f0.mp4", "duration": 181.4729583333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman is flipping through a magazine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman is flipping through a magazine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 36.0], "base_se_norm": [-0.8016233444213867, -0.6032466888427734], "gt_se_raw": [18.0, 36.0], "gt_se_norm": [-0.8016233444213867, -0.6032466888427734], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001901.pt"}
{"idx": 1903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3BCIO1uX6f0.mp4", "duration": 181.4729583333333, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the woman walk over to the man at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the woman walk over to the man at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [-0.4379327893257141, -0.40487003326416016], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.4379327893257141, -0.4158909320831299], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001903.pt"}
{"idx": 1905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HJENMThDg0k.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy open his mouth wide as if in surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy open his mouth wide as if in surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.7814179062843323, -0.6599833965301514], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.8057048320770264, -0.757131040096283], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001905.pt"}
{"idx": 1907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HJENMThDg0k.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy turn his head to the side and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy turn his head to the side and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9271392822265625, -0.8299916982650757], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.392827570438385, -0.2956799864768982], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001907.pt"}
{"idx": 1909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HJENMThDg0k.mp4", "duration": 82.34893333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is touching the boy's shoulders.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is touching the boy's shoulders.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [0.06862342357635498, 0.21434485912322998], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [0.4086400270462036, 0.5057876110076904], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001909.pt"}
{"idx": 1911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4gZpvVa0lis.mp4", "duration": 123.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the chimpanzee is performing sit-ups while the dog holds its feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the chimpanzee is performing sit-ups while the dog holds its feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 58.0], "base_se_norm": [-0.5960140228271484, -0.06275254487991333], "gt_se_raw": [25.0, 55.0], "gt_se_norm": [-0.5960140228271484, -0.11123085021972656], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001911.pt"}
{"idx": 1913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4gZpvVa0lis.mp4", "duration": 123.76666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the chimpanzee cover its face in disappointment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the chimpanzee cover its face in disappointment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [0.6321033239364624, 0.6967411041259766], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [0.6321033239364624, 0.6967411041259766], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001913.pt"}
{"idx": 1915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eRT_mIpXjbs.mp4", "duration": 385.52, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker introduce himself and his role at Kaplan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker introduce himself and his role at Kaplan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9740610122680664, -0.9481220245361328], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.9740610122680664, -0.9533098340034485], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001915.pt"}
{"idx": 1917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eRT_mIpXjbs.mp4", "duration": 385.52, "prompt_text": "Please find the visual event described by the sentence 'At what point is 'Learning Engineering' described as the practical work of applying the science of learning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is 'Learning Engineering' described as the practical work of applying the science of learning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 114.0], "base_se_norm": [-0.6264785528182983, -0.40859097242355347], "gt_se_raw": [71.0, 100.0], "gt_se_norm": [-0.6316663026809692, -0.4812201261520386], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001917.pt"}
{"idx": 1919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eRT_mIpXjbs.mp4", "duration": 385.52, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker talk about the 'gig economy' and its impact on the future of work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 272.0 - 300.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker talk about the 'gig economy' and its impact on the future of work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 272.0 - 300.0 seconds.<|im_end|>\n", "base_se_raw": [272.0, 300.0], "base_se_norm": [0.4110811948776245, 0.5563395023345947], "gt_se_raw": [282.0, 302.0], "gt_se_norm": [0.4629591703414917, 0.5667151212692261], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001919.pt"}
{"idx": 1921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LFoIjKYZtCY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When is there a close-up view of someone welding, creating many sparks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is there a close-up view of someone welding, creating many sparks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001921.pt"}
{"idx": 1923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LFoIjKYZtCY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when four students are posing for the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when four students are posing for the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001923.pt"}
{"idx": 1925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p-_UlScFrQ8.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene with two firefighters in full gear and an instructor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene with two firefighters in full gear and an instructor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001925.pt"}
{"idx": 1927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p-_UlScFrQ8.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the part where an instructor is teaching a group of students in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the part where an instructor is teaching a group of students in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001927.pt"}
{"idx": 1929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p-_UlScFrQ8.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man in a red t-shirt is sitting at a desk in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man in a red t-shirt is sitting at a desk in a classroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001929.pt"}
{"idx": 1931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G9YuKs3Jitk.mp4", "duration": 138.6385, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when hands covered in colorful paint are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when hands covered in colorful paint are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 35.0], "base_se_norm": [-0.5527938008308411, -0.4950897693634033], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.538367748260498, -0.4950897693634033], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001931.pt"}
{"idx": 1933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G9YuKs3Jitk.mp4", "duration": 138.6385, "prompt_text": "Please find the visual event described by the sentence 'At what time does a young boy with glasses wear a funny nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does a young boy with glasses wear a funny nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 62.0], "base_se_norm": [-0.14886558055877686, -0.10558760166168213], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.14886558055877686, -0.10558760166168213], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001933.pt"}
{"idx": 1935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICkM9rvBUzA.mp4", "duration": 135.135, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a waiter serving food in a restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a waiter serving food in a restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.8667998313903809, -0.8075997829437256], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.8519998788833618, -0.8075997829437256], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001935.pt"}
{"idx": 1937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICkM9rvBUzA.mp4", "duration": 135.135, "prompt_text": "Please find the visual event described by the sentence 'At what point do the student and faculty member look at the textbook together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the student and faculty member look at the textbook together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.43759942054748535, -0.3783993721008301], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.43759942054748535, -0.3783993721008301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001937.pt"}
{"idx": 1939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICkM9rvBUzA.mp4", "duration": 135.135, "prompt_text": "Please find the visual event described by the sentence 'When are the women at the registration desk handing out bags to an attendee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the women at the registration desk handing out bags to an attendee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 115.0], "base_se_norm": [0.5540015697479248, 0.7020018100738525], "gt_se_raw": [105.0, 108.0], "gt_se_norm": [0.5540015697479248, 0.598401665687561], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001939.pt"}
{"idx": 1941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFUE09SRO8w.mp4", "duration": 205.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the scene of a storm at a gas station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene of a storm at a gas station.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 71.0], "base_se_norm": [-0.3973756432533264, -0.30989789962768555], "gt_se_raw": [62.0, 71.0], "gt_se_norm": [-0.3973756432533264, -0.30989789962768555], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001941.pt"}
{"idx": 1943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFUE09SRO8w.mp4", "duration": 205.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the animated sequence where a CO2 molecule enters a plant's stoma.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the animated sequence where a CO2 molecule enters a plant's stoma.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 117.0], "base_se_norm": [0.05945253372192383, 0.13721048831939697], "gt_se_raw": [109.0, 117.0], "gt_se_norm": [0.05945253372192383, 0.13721048831939697], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001943.pt"}
{"idx": 1945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMXsAtYtLTo.mp4", "duration": 104.5044, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where farmers are harvesting tomatoes in the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where farmers are harvesting tomatoes in the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.9234482049942017, -0.751206636428833], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.9043102264404297, -0.8277584314346313], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001945.pt"}
{"idx": 1947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMXsAtYtLTo.mp4", "duration": 104.5044, "prompt_text": "Please find the visual event described by the sentence 'At what point are workers manually sorting tomatoes on a production line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are workers manually sorting tomatoes on a production line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.4832753539085388, -0.3875855803489685], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [-0.4641374349594116, -0.4067235589027405], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001947.pt"}
{"idx": 1949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMXsAtYtLTo.mp4", "duration": 104.5044, "prompt_text": "Please find the visual event described by the sentence 'When does the chef start cutting the finished pizza?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the chef start cutting the finished pizza?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 100.0], "base_se_norm": [0.8563811779022217, 0.913794994354248], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [0.703277587890625, 0.7606914043426514], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001949.pt"}
{"idx": 1951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iytz9nbqwi8.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a plaid shirt work on his laptop with a bookshelf behind him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a plaid shirt work on his laptop with a bookshelf behind him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47478586435317993, -0.08087527751922607], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.540437638759613, -0.08087527751922607], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001951.pt"}
{"idx": 1953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iytz9nbqwi8.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where a young woman smiles at the camera in front of many international flags.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where a young woman smiles at the camera in front of many international flags.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4443387985229492, 0.8382494449615479], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3786870241165161, 0.8382494449615479], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001953.pt"}
{"idx": 1955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Juj026QZGDo.mp4", "duration": 284.03375, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a group of people giving each other high fives in a circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a group of people giving each other high fives in a circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.9155030250549316, -0.8802959322929382], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.9155030250549316, -0.8802959322929382], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001955.pt"}
{"idx": 1957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Juj026QZGDo.mp4", "duration": 284.03375, "prompt_text": "Please find the visual event described by the sentence 'When does the woman with pink hair appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman with pink hair appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.8802959322929382, -0.8521302342414856], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.8802959322929382, -0.8521302342414856], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001957.pt"}
{"idx": 1959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Juj026QZGDo.mp4", "duration": 284.03375, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a man in a blue t-shirt with 'I help people get jobs' printed on it is laughing with his friends.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a man in a blue t-shirt with 'I help people get jobs' printed on it is laughing with his friends.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 112.0], "base_se_norm": [-0.23952698707580566, -0.2113613486289978], "gt_se_raw": [108.0, 112.0], "gt_se_norm": [-0.23952698707580566, -0.2113613486289978], "start_token_span": [82, 87], "end_token_span": [89, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001959.pt"}
{"idx": 1961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s1FWVQFeOpQ.mp4", "duration": 130.06326666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in the white t-shirt displaying his phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in the white t-shirt displaying his phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.5848174095153809, -0.5386860370635986], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.5848174095153809, -0.5386860370635986], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001961.pt"}
{"idx": 1963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s1FWVQFeOpQ.mp4", "duration": 130.06326666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man takes a recording device out of his backpack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man takes a recording device out of his backpack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 50.0], "base_se_norm": [-0.292651891708374, -0.23114341497421265], "gt_se_raw": [46.0, 51.0], "gt_se_norm": [-0.292651891708374, -0.21576625108718872], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001963.pt"}
{"idx": 1965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymc30meWzfg.mp4", "duration": 78.74533333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man running on the treadmill while being monitored by a younger man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man running on the treadmill while being monitored by a younger man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.6952199935913086, -0.5682283639907837], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.7206183671951294, -0.5682283639907837], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001965.pt"}
{"idx": 1967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymc30meWzfg.mp4", "duration": 78.74533333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman in a patterned blouse is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman in a patterned blouse is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.6698216795921326, -0.49203336238861084], "gt_se_raw": [61.0, 69.0], "gt_se_norm": [0.5492981672286987, 0.752484917640686], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001967.pt"}
{"idx": 1969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymc30meWzfg.mp4", "duration": 78.74533333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a group of people is observing a man walking on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a group of people is observing a man walking on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [-0.49203336238861084, -0.33964335918426514], "gt_se_raw": [20.0, 26.0], "gt_se_norm": [-0.49203336238861084, -0.33964335918426514], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001969.pt"}
{"idx": 1971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YOqmroV2cRo.mp4", "duration": 66.7667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man and the child are interacting closely at the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man and the child are interacting closely at the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 66.0], "base_se_norm": [-0.5506742000579834, 0.9770334959030151], "gt_se_raw": [15.0, 62.0], "gt_se_norm": [-0.5506742000579834, 0.8572132587432861], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001971.pt"}
{"idx": 1973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YOqmroV2cRo.mp4", "duration": 66.7667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man holds the child's hand and starts laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man holds the child's hand and starts laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 66.0], "base_se_norm": [0.3180222511291504, 0.9770334959030151], "gt_se_raw": [57.0, 62.0], "gt_se_norm": [0.7074379920959473, 0.8572132587432861], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001973.pt"}
{"idx": 1975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSjK0X2XLxY.mp4", "duration": 191.52485819152486, "prompt_text": "Please find the visual event described by the sentence 'At what point are all the necessary materials for the experiment displayed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are all the necessary materials for the experiment displayed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 60.0], "base_se_norm": [-0.7284947633743286, -0.373449444770813], "gt_se_raw": [26.0, 60.0], "gt_se_norm": [-0.7284947633743286, -0.373449444770813], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001975.pt"}
{"idx": 1977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSjK0X2XLxY.mp4", "duration": 191.52485819152486, "prompt_text": "Please find the visual event described by the sentence 'When is a person using a straw to blow a bubble into the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person using a straw to blow a bubble into the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 116.0], "base_se_norm": [0.1277909278869629, 0.21133100986480713], "gt_se_raw": [109.0, 116.0], "gt_se_norm": [0.13823342323303223, 0.21133100986480713], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001977.pt"}
{"idx": 1979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xSjK0X2XLxY.mp4", "duration": 191.52485819152486, "prompt_text": "Please find the visual event described by the sentence 'At what time is the final photo art of the bubble displayed in a white frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 174.0 - 178.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the final photo art of the bubble displayed in a white frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 174.0 - 178.0 seconds.<|im_end|>\n", "base_se_raw": [174.0, 178.0], "base_se_norm": [0.8169965744018555, 0.8587665557861328], "gt_se_raw": [175.0, 179.0], "gt_se_norm": [0.8274390697479248, 0.8692090511322021], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001979.pt"}
{"idx": 1981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM-e5QkRNo.mp4", "duration": 27.39267296229322, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the dog's head aligns with the hanging toys, making them look like ears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the dog's head aligns with the hanging toys, making them look like ears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.8539755344390869, -0.5619266629219055], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9269877672195435, -0.41590219736099243], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001981.pt"}
{"idx": 1983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM-e5QkRNo.mp4", "duration": 27.39267296229322, "prompt_text": "Please find the visual event described by the sentence 'Show me when the dog is standing calmly behind the gate and panting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the dog is standing calmly behind the gate and panting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [0.6062688827514648, 0.9713300466537476], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [0.6062688827514648, 0.9713300466537476], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001983.pt"}
{"idx": 1985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r07eYlcQLJQ.mp4", "duration": 70.1, "prompt_text": "Please find the visual event described by the sentence 'When does the dog emerge from beneath the trampoline and walk on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog emerge from beneath the trampoline and walk on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.6576319336891174, -0.48644793033599854], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.6291012763977051, -0.5149785876274109], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001985.pt"}
{"idx": 1987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r07eYlcQLJQ.mp4", "duration": 70.1, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see a second dog standing under the trees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see a second dog standing under the trees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.20114123821258545, -0.05848783254623413], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.14407986402511597, -0.05848783254623413], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001987.pt"}
{"idx": 1989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r07eYlcQLJQ.mp4", "duration": 70.1, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time range when the drooping sunflowers in the garden are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time range when the drooping sunflowers in the garden are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 63.0], "base_se_norm": [0.34094154834747314, 0.7974323034286499], "gt_se_raw": [54.0, 62.0], "gt_se_norm": [0.5406562089920044, 0.7689015865325928], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001989.pt"}
{"idx": 1991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1ycCh8MGLDo.mp4", "duration": 269.72779166666663, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the person is boiling the tea leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the person is boiling the tea leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 90.0], "base_se_norm": [-0.40680932998657227, -0.3326604962348938], "gt_se_raw": [81.0, 89.0], "gt_se_norm": [-0.3993944525718689, -0.34007537364959717], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001991.pt"}
{"idx": 1993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1ycCh8MGLDo.mp4", "duration": 269.72779166666663, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for straining the milk tea into the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for straining the milk tea into the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 158.0], "base_se_norm": [-0.058309853076934814, 0.17155158519744873], "gt_se_raw": [129.0, 143.0], "gt_se_norm": [-0.043480098247528076, 0.06032824516296387], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001993.pt"}
{"idx": 1995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vY0I6yjBy1c.mp4", "duration": 92.42566666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman demonstrate the toilet paper shortage by stacking rolls on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman demonstrate the toilet paper shortage by stacking rolls on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 36.0], "base_se_norm": [-0.4373857378959656, -0.22099560499191284], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.41574668884277344, -0.22099560499191284], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001995.pt"}
{"idx": 1997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vY0I6yjBy1c.mp4", "duration": 92.42566666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the blue striped shirt start to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the blue striped shirt start to speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 58.0], "base_se_norm": [0.038672566413879395, 0.2550625801086426], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.038672566413879395, 0.12522852420806885], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001997.pt"}
{"idx": 1999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vY0I6yjBy1c.mp4", "duration": 92.42566666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the low-angle shot from a bare shelf looking towards a shopper in the distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the low-angle shot from a bare shelf looking towards a shopper in the distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 80.0], "base_se_norm": [0.601286768913269, 0.7311208248138428], "gt_se_raw": [73.0, 80.0], "gt_se_norm": [0.5796477794647217, 0.7311208248138428], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00001999.pt"}
{"idx": 2001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4ClKFnnzSRA.mp4", "duration": 134.86806666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in purple holds up a white paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in purple holds up a white paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [-0.8665362596511841, -0.7775604128837585], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.8517069220542908, -0.8220483064651489], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002001.pt"}
{"idx": 2003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4ClKFnnzSRA.mp4", "duration": 134.86806666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the purple shirt lean in to talk to the person in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the purple shirt lean in to talk to the person in the wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9555121064186096, -0.8665362596511841], "gt_se_raw": [99.0, 102.0], "gt_se_norm": [0.4681013822555542, 0.5125892162322998], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002003.pt"}
{"idx": 2005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mnd7sFt5c3A.mp4", "duration": 156.82333333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the boy load a rifle in the shed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy load a rifle in the shed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [-0.4516122341156006, -0.4005994200706482], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.4516122341156006, -0.4005994200706482], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002005.pt"}
{"idx": 2007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mnd7sFt5c3A.mp4", "duration": 156.82333333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point do the police officer and the worried woman have a conversation in an office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the police officer and the worried woman have a conversation in an office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 62.0], "base_se_norm": [-0.7066763043403625, -0.20930135250091553], "gt_se_raw": [49.0, 62.0], "gt_se_norm": [-0.3750929832458496, -0.20930135250091553], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002007.pt"}
{"idx": 2009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mnd7sFt5c3A.mp4", "duration": 156.82333333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the lightbulb in the shed explodes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the lightbulb in the shed explodes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [-0.19654816389083862, -0.14553529024124146], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [-0.20930135250091553, -0.17104172706604004], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002009.pt"}
{"idx": 2011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V49ENdZlOx4.mp4", "duration": 30.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the minion successfully throws the crumpled paper into the trash can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the minion successfully throws the crumpled paper into the trash can.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [0.6000000238418579, 0.7333333492279053], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.7333333492279053, -0.6000000238418579], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002011.pt"}
{"idx": 2013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V49ENdZlOx4.mp4", "duration": 30.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the crumpled paper ball is seen floating upwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the crumpled paper ball is seen floating upwards.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [0.6000000238418579, 0.7333333492279053], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.6000000238418579, 0.7333333492279053], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002013.pt"}
{"idx": 2015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/duF40iZq464.mp4", "duration": 134.50103333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a researcher walking in the woods with a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a researcher walking in the woods with a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 31.0], "base_se_norm": [-0.7472138404846191, -0.5390369892120361], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.6133859157562256, -0.5390369892120361], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002015.pt"}
{"idx": 2017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/duF40iZq464.mp4", "duration": 134.50103333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two chipmunks are foraging on the ground together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two chipmunks are foraging on the ground together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 63.0], "base_se_norm": [-0.10781359672546387, -0.06320422887802124], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [-0.09294378757476807, -0.06320422887802124], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002017.pt"}
{"idx": 2019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/duF40iZq464.mp4", "duration": 134.50103333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the chipmunk with a blue collar appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the chipmunk with a blue collar appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 72.0], "base_se_norm": [0.011144638061523438, 0.07062375545501709], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [0.02601444721221924, 0.07062375545501709], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002019.pt"}
{"idx": 2021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-cf_-i_gCdY.mp4", "duration": 138.5, "prompt_text": "Please find the visual event described by the sentence 'A man in a gray hoodie falls asleep on a kitchen counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a gray hoodie falls asleep on a kitchen counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.6534296274185181, -0.5667870044708252], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [-0.6245487332344055, -0.5667870044708252], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002021.pt"}
{"idx": 2023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-cf_-i_gCdY.mp4", "duration": 138.5, "prompt_text": "Please find the visual event described by the sentence 'A drawing of two Native Americans smoking a peace pipe is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A drawing of two Native Americans smoking a peace pipe is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.32129961252212524, -0.24909746646881104], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [-0.32129961252212524, -0.24909746646881104], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002023.pt"}
{"idx": 2025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8eBmqNyxBg.mp4", "duration": 117.7176, "prompt_text": "Please find the visual event described by the sentence 'When does the woman pluck a feather from the chicken's body?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman pluck a feather from the chicken's body?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.5922445058822632, -0.5412750244140625], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.5752546787261963, -0.5242851972579956], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002025.pt"}
{"idx": 2027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8eBmqNyxBg.mp4", "duration": 117.7176, "prompt_text": "Please find the visual event described by the sentence 'At what point does the chicken hide under the buckets and shuffle them around?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the chicken hide under the buckets and shuffle them around?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 73.0], "base_se_norm": [0.0703582763671875, 0.24025630950927734], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [0.17229712009429932, 0.2232666015625], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002027.pt"}
{"idx": 2029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8eBmqNyxBg.mp4", "duration": 117.7176, "prompt_text": "Please find the visual event described by the sentence 'When does the woman find out her baby is missing from the crib?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman find out her baby is missing from the crib?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 97.0], "base_se_norm": [0.5800526142120361, 0.6480119228363037], "gt_se_raw": [94.0, 97.0], "gt_se_norm": [0.597042441368103, 0.6480119228363037], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002029.pt"}
{"idx": 2031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GoDvDNGaPUU.mp4", "duration": 15.081733333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the parrot start preening its feathers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the parrot start preening its feathers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.469556987285614, 0.7239397764205933], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.3369462490081787, 0.06088602542877197], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002031.pt"}
{"idx": 2033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GoDvDNGaPUU.mp4", "duration": 15.081733333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the parrot briefly stops grooming and looks towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the parrot briefly stops grooming and looks towards the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 14.0], "base_se_norm": [0.7239397764205933, 0.8565504550933838], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [0.7239397764205933, 0.8565504550933838], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002033.pt"}
{"idx": 2035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kO_mpRhI1jY.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we observe two men working on server racks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe two men working on server racks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.540437638759613], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002035.pt"}
{"idx": 2037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kO_mpRhI1jY.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a man is using a screwdriver to repair a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a man is using a screwdriver to repair a laptop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.08087527751922607, 0.3786870241165161], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08087527751922607, 0.3786870241165161], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002037.pt"}
{"idx": 2039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kO_mpRhI1jY.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.05042827129364014, 0.3786870241165161], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002039.pt"}
{"idx": 2041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UkcYowvTRdA.mp4", "duration": 55.7, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman with curly red hair and glasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman with curly red hair and glasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 22.0], "base_se_norm": [-0.9281867146492004, -0.21005386114120483], "gt_se_raw": [3.0, 23.0], "gt_se_norm": [-0.892280101776123, -0.17414724826812744], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002041.pt"}
{"idx": 2043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UkcYowvTRdA.mp4", "duration": 55.7, "prompt_text": "Please find the visual event described by the sentence 'At what point does the subtitle 'All are based on Abator's passion for helping people achieve intelligent use of information!' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the subtitle 'All are based on Abator's passion for helping people achieve intelligent use of information!' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [0.5798922777175903, 0.7953320741653442], "gt_se_raw": [44.0, 50.0], "gt_se_norm": [0.5798922777175903, 0.7953320741653442], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002043.pt"}
{"idx": 2045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FeaDpI96lIA.mp4", "duration": 67.20045045045045, "prompt_text": "Please find the visual event described by the sentence 'When are two women sitting at a table having a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two women sitting at a table having a conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.8511914610862732, -0.7619063258171082], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.8809531927108765, -0.7916680574417114], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002045.pt"}
{"idx": 2047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FeaDpI96lIA.mp4", "duration": 67.20045045045045, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman is standing next to a large sign for the CSU Health and Medical Center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman is standing next to a large sign for the CSU Health and Medical Center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.25595730543136597, -0.1666722297668457], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [-0.28571903705596924, -0.1666722297668457], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002047.pt"}
{"idx": 2049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FeaDpI96lIA.mp4", "duration": 67.20045045045045, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman is holding a white measuring spoon with some ingredients for another person to see.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman is holding a white measuring spoon with some ingredients for another person to see.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.345242440700531, -0.25595730543136597], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.3750041723251343, -0.28571903705596924], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002049.pt"}
{"idx": 2051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VqrBsMFRaLA.mp4", "duration": 203.11957777777778, "prompt_text": "Please find the visual event described by the sentence 'Show the woman wearing a white fur coat holding up a newspaper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman wearing a white fur coat holding up a newspaper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.5175255537033081, -0.48798632621765137], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.5175255537033081, -0.48798632621765137], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002051.pt"}
{"idx": 2053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VqrBsMFRaLA.mp4", "duration": 203.11957777777778, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman with green skin and black horns is singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman with green skin and black horns is singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 22.0], "base_se_norm": [-0.9507679343223572, -0.7833788394927979], "gt_se_raw": [89.0, 91.0], "gt_se_norm": [-0.12366890907287598, -0.10397613048553467], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002053.pt"}
{"idx": 2055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KHHrCdQZ5dk.mp4", "duration": 30.2, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in pink slide down the red ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in pink slide down the red ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.2715231776237488, -0.07284772396087646], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.2715231776237488, -0.07284772396087646], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002055.pt"}
{"idx": 2057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KHHrCdQZ5dk.mp4", "duration": 30.2, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the instructor in the orange shirt helps the girl position herself at the top of the ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the instructor in the orange shirt helps the girl position herself at the top of the ramp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 12.0], "base_se_norm": [-0.8013244867324829, -0.20529800653457642], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.7350993156433105, -0.2715231776237488], "start_token_span": [78, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002057.pt"}
{"idx": 2059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KHHrCdQZ5dk.mp4", "duration": 30.2, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the girl in pink is crawling on the blue mat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the girl in pink is crawling on the blue mat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 19.0], "base_se_norm": [-0.13907289505004883, 0.25827813148498535], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.13907289505004883, 0.05960261821746826], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002059.pt"}
{"idx": 2061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vKtnj21k1Cg.mp4", "duration": 129.7296, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in blue scrubs use sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in blue scrubs use sign language?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 44.0], "base_se_norm": [-0.5374995470046997, -0.32166600227355957], "gt_se_raw": [31.0, 40.0], "gt_se_norm": [-0.5220828652381897, -0.3833327293395996], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002061.pt"}
{"idx": 2063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vKtnj21k1Cg.mp4", "duration": 129.7296, "prompt_text": "Please find the visual event described by the sentence 'Show me when a person is typing on a slide-out keyboard of a mobile phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a person is typing on a slide-out keyboard of a mobile phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 91.0], "base_se_norm": [0.35666799545288086, 0.40291810035705566], "gt_se_raw": [88.0, 91.0], "gt_se_norm": [0.35666799545288086, 0.40291810035705566], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002063.pt"}
{"idx": 2065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gRD53bcAM8E.mp4", "duration": 52.11873333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the brown bear pull a banana out of its ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the brown bear pull a banana out of its ear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [-0.5011390447616577, -0.4243912100791931], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.5395129919052124, -0.4627651572227478], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002065.pt"}
{"idx": 2067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gRD53bcAM8E.mp4", "duration": 52.11873333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a chef puppet with a microwave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a chef puppet with a microwave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.11739987134933472, 0.0744696855545044], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.1557738184928894, 0.0744696855545044], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002067.pt"}
{"idx": 2069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gRD53bcAM8E.mp4", "duration": 52.11873333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the popcorn fly into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the popcorn fly into the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [0.2279653549194336, 0.30471324920654297], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [0.2279653549194336, 0.30471324920654297], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002069.pt"}
{"idx": 2071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7jck10cm5g.mp4", "duration": 98.16473333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are hands shown operating a Perkins Brailler typewriter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are hands shown operating a Perkins Brailler typewriter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.714765191078186, -0.572147786617279], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.02205204963684082, 0.05944359302520752], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002071.pt"}
{"idx": 2073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7jck10cm5g.mp4", "duration": 98.16473333333333, "prompt_text": "Please find the visual event described by the sentence 'When does a hand open a wooden card catalog drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a hand open a wooden card catalog drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 97.0], "base_se_norm": [0.9355219602584839, 0.9762698411941528], "gt_se_raw": [95.0, 98.0], "gt_se_norm": [0.9355219602584839, 0.9966437816619873], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002073.pt"}
{"idx": 2075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGeKSiCQkPw.mp4", "duration": 80.6806, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video does the dog have its mouth wide open, looking very surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video does the dog have its mouth wide open, looking very surprised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 45.0], "base_se_norm": [0.06593155860900879, 0.1155097484588623], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.33069413900375366, -0.28111594915390015], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002075.pt"}
{"idx": 2077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGeKSiCQkPw.mp4", "duration": 80.6806, "prompt_text": "Please find the visual event described by the sentence 'When can we see the dog yawning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the dog yawning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 45.0], "base_se_norm": [0.06593155860900879, 0.1155097484588623], "gt_se_raw": [43.0, 44.0], "gt_se_norm": [0.06593155860900879, 0.09072065353393555], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002077.pt"}
{"idx": 2079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGeKSiCQkPw.mp4", "duration": 80.6806, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the dog lets out a final, loud scream.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the dog lets out a final, loud scream.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [0.561713695526123, 0.6360809803009033], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [0.5865027904510498, 0.6360809803009033], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002079.pt"}
{"idx": 2081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wkbyROy79oY.mp4", "duration": 107.2071, "prompt_text": "Please find the visual event described by the sentence 'Show me the photo where a boy is wearing large yellow goggles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the photo where a boy is wearing large yellow goggles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.32840269804000854, -0.2537807822227478], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [-0.34705817699432373, -0.2537807822227478], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002081.pt"}
{"idx": 2083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wkbyROy79oY.mp4", "duration": 107.2071, "prompt_text": "Please find the visual event described by the sentence 'When does a boy in a hallway play with a small helicopter toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a boy in a hallway play with a small helicopter toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 95.0], "base_se_norm": [0.6976487636566162, 0.772270679473877], "gt_se_raw": [91.0, 96.0], "gt_se_norm": [0.6976487636566162, 0.7909262180328369], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002083.pt"}
{"idx": 2085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ocAL0oz3HxU.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'When is the boy lying on the floor and playing with a yellow cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy lying on the floor and playing with a yellow cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 25.0], "base_se_norm": [-0.9403937458992004, -0.7019686698913574], "gt_se_raw": [5.0, 25.0], "gt_se_norm": [-0.9403937458992004, -0.7019686698913574], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002085.pt"}
{"idx": 2087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ocAL0oz3HxU.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point is the boy sitting in a space with colorful lights on the ceiling and a dark blue backboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the boy sitting in a space with colorful lights on the ceiling and a dark blue backboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 78.0], "base_se_norm": [-0.23703980445861816, -0.07014220952987671], "gt_se_raw": [63.0, 79.0], "gt_se_norm": [-0.24896103143692017, -0.05822098255157471], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002087.pt"}
{"idx": 2089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ocAL0oz3HxU.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'When is a boy sitting in a wheelchair while an adult interacts with him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a boy sitting in a wheelchair while an adult interacts with him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 121.0], "base_se_norm": [0.335180401802063, 0.44247162342071533], "gt_se_raw": [116.0, 121.0], "gt_se_norm": [0.38286542892456055, 0.44247162342071533], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002089.pt"}
{"idx": 2091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0x1EwpvvyeQ.mp4", "duration": 120.38693333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is a list of disabilities shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is a list of disabilities shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 43.0], "base_se_norm": [-0.4185415506362915, -0.2856367230415344], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [-0.4185415506362915, -0.2856367230415344], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002091.pt"}
{"idx": 2093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0x1EwpvvyeQ.mp4", "duration": 120.38693333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a person in a wheelchair is playing a game on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a person in a wheelchair is playing a game on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 55.0], "base_se_norm": [-0.25241053104400635, -0.08627957105636597], "gt_se_raw": [46.0, 55.0], "gt_se_norm": [-0.2357974648475647, -0.08627957105636597], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002093.pt"}
{"idx": 2095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L9KC5W7A2yI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a student working on a furnace while being supervised by two instructors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a student working on a furnace while being supervised by two instructors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002095.pt"}
{"idx": 2097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L9KC5W7A2yI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the close-up shot of hands adjusting a manifold gauge set?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the close-up shot of hands adjusting a manifold gauge set?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.08188092708587646, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002097.pt"}
{"idx": 2099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L9KC5W7A2yI.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.049278974533081055, 0.3771786689758301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002099.pt"}
{"idx": 2101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfyJQEIsMt0.mp4", "duration": 319.3607083333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the stork get bitten by the baby alligator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the stork get bitten by the baby alligator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 83.0], "base_se_norm": [-0.49899911880493164, -0.48021161556243896], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [-0.44263654947280884, -0.4238489866256714], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002101.pt"}
{"idx": 2103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfyJQEIsMt0.mp4", "duration": 319.3607083333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the grey cloud is crying and it starts to rain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 232.0 - 237.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the grey cloud is crying and it starts to rain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 232.0 - 237.0 seconds.<|im_end|>\n", "base_se_raw": [232.0, 237.0], "base_se_norm": [0.45290255546569824, 0.4842151403427124], "gt_se_raw": [233.0, 237.0], "gt_se_norm": [0.459165096282959, 0.4842151403427124], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002103.pt"}
{"idx": 2105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FkTybqcX-Yo.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'When does the man say 'You're a wanted man'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say 'You're a wanted man'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.7954264283180237, -0.7545117139816284], "gt_se_raw": [21.0, 22.0], "gt_se_norm": [-0.7135969400405884, -0.6999587416648865], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002105.pt"}
{"idx": 2107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FkTybqcX-Yo.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the woman is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the woman is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [-0.07259976863861084, -0.031685054302215576], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.07259976863861084, -0.04532325267791748], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002107.pt"}
{"idx": 2109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FkTybqcX-Yo.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'Show the fight scene between Iron Man, Captain America, and Bucky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the fight scene between Iron Man, Captain America, and Bucky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 136.0], "base_se_norm": [0.7729710340499878, 0.8548004627227783], "gt_se_raw": [130.0, 136.0], "gt_se_norm": [0.7729710340499878, 0.8548004627227783], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002109.pt"}
{"idx": 2111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GaoLU6zKaws.mp4", "duration": 302.5, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the musician is playing his instrument while inside a shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the musician is playing his instrument while inside a shopping cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 111.0], "base_se_norm": [-0.2991735339164734, -0.2661157250404358], "gt_se_raw": [103.0, 106.0], "gt_se_norm": [-0.3190082907676697, -0.2991735339164734], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002111.pt"}
{"idx": 2113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GaoLU6zKaws.mp4", "duration": 302.5, "prompt_text": "Please find the visual event described by the sentence 'When do we see a man in a red jacket looking surprised at the saxophone player in the food court?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see a man in a red jacket looking surprised at the saxophone player in the food court?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 46.0], "base_se_norm": [-0.7090909481048584, -0.6958677768707275], "gt_se_raw": [43.0, 46.0], "gt_se_norm": [-0.715702474117279, -0.6958677768707275], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002113.pt"}
{"idx": 2115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vjW4UOC7U3w.mp4", "duration": 73.30656666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a black jacket is working with the wires of a stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a black jacket is working with the wires of a stove.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.7271731495857239, -0.6180423498153687], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.7271731495857239, -0.6180423498153687], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002115.pt"}
{"idx": 2117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vjW4UOC7U3w.mp4", "duration": 73.30656666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man uses a screwdriver to fix a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man uses a screwdriver to fix a machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8635865449905396, -0.7271731495857239], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.8635865449905396, -0.7271731495857239], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002117.pt"}
{"idx": 2119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vjW4UOC7U3w.mp4", "duration": 73.30656666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man in a red shirt is happily playing with a ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man in a red shirt is happily playing with a ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 72.0], "base_se_norm": [0.9097880125045776, 0.9643534421920776], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [0.9097880125045776, 0.9643534421920776], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002119.pt"}
{"idx": 2121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2PjZAeiU7uM.mp4", "duration": 154.98816666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in a black turtleneck and glasses holding a red box and pointing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in a black turtleneck and glasses holding a red box and pointing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.8193410634994507, -0.7548199892044067], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [-0.8064368367195129, -0.7935326099395752], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002121.pt"}
{"idx": 2123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2PjZAeiU7uM.mp4", "duration": 154.98816666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman changes from clogs into high heels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman changes from clogs into high heels.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.19993889331817627, -0.1612262725830078], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [-0.19993889331817627, -0.17413043975830078], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002123.pt"}
{"idx": 2125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YipS4o9EaoU.mp4", "duration": 158.65849182515848, "prompt_text": "Please find the visual event described by the sentence 'When is the slide explaining the 'Severity' of the illness shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the slide explaining the 'Severity' of the illness shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 40.0], "base_se_norm": [-0.6218292713165283, -0.4957723617553711], "gt_se_raw": [30.0, 41.0], "gt_se_norm": [-0.6218292713165283, -0.4831666350364685], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002125.pt"}
{"idx": 2127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YipS4o9EaoU.mp4", "duration": 158.65849182515848, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker make a gesture with both hands, showing two fingers on each?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker make a gesture with both hands, showing two fingers on each?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 78.0], "base_se_norm": [-0.054573118686676025, -0.016756057739257812], "gt_se_raw": [76.0, 77.0], "gt_se_norm": [-0.04196745157241821, -0.0293617844581604], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002127.pt"}
{"idx": 2129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YipS4o9EaoU.mp4", "duration": 158.65849182515848, "prompt_text": "Please find the visual event described by the sentence 'Find the segment that shows a collection of shoes laid out on the grass and pavement as part of a protest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment that shows a collection of shoes laid out on the grass and pavement as part of a protest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 143.0], "base_se_norm": [0.7395853996276855, 0.8026138544082642], "gt_se_raw": [137.0, 144.0], "gt_se_norm": [0.7269797325134277, 0.815219521522522], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002129.pt"}
{"idx": 2131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MCKeh7ynQjQ.mp4", "duration": 98.01458333333332, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when she discusses the results of the survey conducted among graduates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when she discusses the results of the survey conducted among graduates.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 57.0], "base_se_norm": [-0.32663077116012573, 0.1630922555923462], "gt_se_raw": [40.0, 50.0], "gt_se_norm": [-0.18379491567611694, 0.020256400108337402], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002131.pt"}
{"idx": 2133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MCKeh7ynQjQ.mp4", "duration": 98.01458333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker advise graduates to stay safe and listen to healthcare experts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker advise graduates to stay safe and listen to healthcare experts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 57.0], "base_se_norm": [-0.12257951498031616, 0.1630922555923462], "gt_se_raw": [69.0, 77.0], "gt_se_norm": [0.4079538583755493, 0.5711947679519653], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002133.pt"}
{"idx": 2135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1DDhUnyvwY.mp4", "duration": 162.6, "prompt_text": "Please find the visual event described by the sentence 'When can we see a monkey catching insects on top of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a monkey catching insects on top of a tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 71.0], "base_se_norm": [-0.24969249963760376, -0.1266912817955017], "gt_se_raw": [65.0, 71.0], "gt_se_norm": [-0.2004920244216919, -0.1266912817955017], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002135.pt"}
{"idx": 2137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1DDhUnyvwY.mp4", "duration": 162.6, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a baby chimpanzee lying on the ground being groomed by its mother?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a baby chimpanzee lying on the ground being groomed by its mother?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 84.0], "base_se_norm": [-0.02829033136367798, 0.03321027755737305], "gt_se_raw": [79.0, 84.0], "gt_se_norm": [-0.02829033136367798, 0.03321027755737305], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002137.pt"}
{"idx": 2139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1DDhUnyvwY.mp4", "duration": 162.6, "prompt_text": "Please find the visual event described by the sentence 'Can you locate the clip of the earth zooming in on Africa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you locate the clip of the earth zooming in on Africa?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9630996584892273, -0.9261992573738098], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9630996584892273, -0.9138991236686707], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002139.pt"}
{"idx": 2141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iRsB9WrSePA.mp4", "duration": 75.44210877544211, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of medical staff at an outdoor drive-thru testing site.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of medical staff at an outdoor drive-thru testing site.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 45.0], "base_se_norm": [0.007394909858703613, 0.19296777248382568], "gt_se_raw": [38.0, 45.0], "gt_se_norm": [0.007394909858703613, 0.19296777248382568], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002141.pt"}
{"idx": 2143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iRsB9WrSePA.mp4", "duration": 75.44210877544211, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a green shirt speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a green shirt speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 38.0], "base_se_norm": [-0.15166741609573364, 0.007394909858703613], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.15166741609573364, 0.007394909858703613], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002143.pt"}
{"idx": 2145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SY0VyoUP5uw.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When can we see the employment rate for non-disabled Americans in 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the employment rate for non-disabled Americans in 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [78, 81], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002145.pt"}
{"idx": 2147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SY0VyoUP5uw.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the narrator says 'employed at nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the narrator says 'employed at nearly half the rate'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.030096113681793213, 0.16388463973999023], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.030096113681793213, 0.16388463973999023], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002147.pt"}
{"idx": 2149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SY0VyoUP5uw.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention that the employment rate for Americans without disabilities is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention that the employment rate for Americans without disabilities is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9030095934867859, -0.4180576801300049], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9030095934867859, -0.6120384335517883], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002149.pt"}
{"idx": 2151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OzP8k0R-USw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the two men walk out of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the two men walk out of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.6907117366790771, -0.6535971760749817], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7030832767486572, -0.6659687161445618], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002151.pt"}
{"idx": 2153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OzP8k0R-USw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman carried into the gallery on her bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman carried into the gallery on her bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 103.0], "base_se_norm": [0.2124098539352417, 0.2742675542831421], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.2124098539352417, 0.261896014213562], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002153.pt"}
{"idx": 2155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MrFs4lL2pbc.mp4", "duration": 91.32456666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two children lying on the floor and playing a hand game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two children lying on the floor and playing a hand game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 91.0], "base_se_norm": [-1.0, 0.9928919076919556], "gt_se_raw": [0.0, 91.0], "gt_se_norm": [-1.0, 0.9928919076919556], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002155.pt"}
{"idx": 2157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MrFs4lL2pbc.mp4", "duration": 91.32456666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the children's hands form a pyramid shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the children's hands form a pyramid shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 61.0], "base_se_norm": [0.18259525299072266, 0.3358945846557617], "gt_se_raw": [54.0, 60.0], "gt_se_norm": [0.18259525299072266, 0.3139946460723877], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002157.pt"}
{"idx": 2159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MrFs4lL2pbc.mp4", "duration": 91.32456666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy wear a baseball cap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy wear a baseball cap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 90.0], "base_se_norm": [-1.0, 0.9709920883178711], "gt_se_raw": [0.0, 91.0], "gt_se_norm": [-1.0, 0.9928919076919556], "start_token_span": [69, 72], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002159.pt"}
{"idx": 2161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzaVd6zl2bA.mp4", "duration": 106.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man in the silver shirt enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man in the silver shirt enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.5871129035949707, -0.5120425224304199], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [-0.5871129035949707, -0.5120425224304199], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002161.pt"}
{"idx": 2163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzaVd6zl2bA.mp4", "duration": 106.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man with white paint on his nose is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man with white paint on his nose is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [0.06975293159484863, 0.12605571746826172], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [0.06975293159484863, 0.12605571746826172], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002163.pt"}
{"idx": 2165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oK6uI4FzYkg.mp4", "duration": 59.934875, "prompt_text": "Please find the visual event described by the sentence 'Show me the exterior shot of the building against a cloudy blue sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the exterior shot of the building against a cloudy blue sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [0.23467350006103516, 0.4015212059020996], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [0.23467350006103516, 0.4015212059020996], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002165.pt"}
{"idx": 2167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oK6uI4FzYkg.mp4", "duration": 59.934875, "prompt_text": "Please find the visual event described by the sentence 'A person walks past several tables of students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person walks past several tables of students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [0.6351081132888794, 0.8686950206756592], "gt_se_raw": [49.0, 57.0], "gt_se_norm": [0.6351081132888794, 0.9020645618438721], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002167.pt"}
{"idx": 2169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oK6uI4FzYkg.mp4", "duration": 59.934875, "prompt_text": "Please find the visual event described by the sentence 'Point to the moment where the camera pans upwards to show the circular light fixtures.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Point to the moment where the camera pans upwards to show the circular light fixtures.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.365978479385376, -0.23250025510787964], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.3326089382171631, -0.23250025510787964], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002169.pt"}
{"idx": 2171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kvt41MIelDE.mp4", "duration": 105.35524444444445, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the green kite is flying down the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the green kite is flying down the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 38.0], "base_se_norm": [-0.46846503019332886, -0.2786310911178589], "gt_se_raw": [28.0, 35.0], "gt_se_norm": [-0.46846503019332886, -0.3355812430381775], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002171.pt"}
{"idx": 2173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kvt41MIelDE.mp4", "duration": 105.35524444444445, "prompt_text": "Please find the visual event described by the sentence 'Show me the time range when a young boy is running across a grassy field with a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time range when a young boy is running across a grassy field with a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 61.0], "base_se_norm": [-0.0698137879371643, 0.15798687934875488], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [0.08205342292785645, 0.15798687934875488], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002173.pt"}
{"idx": 2175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJosiEHwWxc.mp4", "duration": 78.91216666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the yellow shirt enter the restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the yellow shirt enter the restaurant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.923966109752655, -0.8479322195053101], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.9493107199668884, -0.8225875496864319], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002175.pt"}
{"idx": 2177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJosiEHwWxc.mp4", "duration": 78.91216666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man in the yellow shirt walking towards the play area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man in the yellow shirt walking towards the play area?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 40.0], "base_se_norm": [-0.44241809844970703, 0.013785362243652344], "gt_se_raw": [22.0, 33.0], "gt_se_norm": [-0.44241809844970703, -0.16362708806991577], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002177.pt"}
{"idx": 2179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LJosiEHwWxc.mp4", "duration": 78.91216666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where the man in the yellow shirt is sitting at a table with his meal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where the man in the yellow shirt is sitting at a table with his meal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.7718982696533203, -0.594485878944397], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.7212090492248535, -0.594485878944397], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002179.pt"}
{"idx": 2181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yBiV0NOeXIY.mp4", "duration": 39.6396, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video become blurry to simulate poor visual acuity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video become blurry to simulate poor visual acuity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 16.0], "base_se_norm": [-0.49545401334762573, -0.19272643327713013], "gt_se_raw": [10.0, 17.0], "gt_se_norm": [-0.49545401334762573, -0.14227181673049927], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002181.pt"}
{"idx": 2183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yBiV0NOeXIY.mp4", "duration": 39.6396, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman points her index finger forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman points her index finger forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9495453834533691, -0.7981815934181213], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.6972724199295044, -0.5963631868362427], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002183.pt"}
{"idx": 2185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/usnrNhN_eUA.mp4", "duration": 102.43566666666666, "prompt_text": "Please find the visual event described by the sentence 'When do we see people standing in line for food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see people standing in line for food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.37521761655807495, -0.33616870641708374], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.37521761655807495, -0.33616870641708374], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002185.pt"}
{"idx": 2187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/usnrNhN_eUA.mp4", "duration": 102.43566666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where an elderly woman is learning to use an electronic reading aid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where an elderly woman is learning to use an electronic reading aid.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.2580709457397461, -0.17997312545776367], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.2385464906692505, -0.17997312545776367], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002187.pt"}
{"idx": 2189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/usnrNhN_eUA.mp4", "duration": 102.43566666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time is a man shown playing the harp for an audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a man shown playing the harp for an audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 61.0], "base_se_norm": [-0.023777544498443604, 0.19099140167236328], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [0.13241803646087646, 0.17146694660186768], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002189.pt"}
{"idx": 2191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-FcsIyqJDzc.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the puppy climbing on top of the sleeping adult dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the puppy climbing on top of the sleeping adult dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 24.0], "base_se_norm": [-0.4447871446609497, -0.1116594672203064], "gt_se_raw": [16.0, 27.0], "gt_se_norm": [-0.407772958278656, -0.000616908073425293], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002191.pt"}
{"idx": 2193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-FcsIyqJDzc.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the puppy face the camera and lick its nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the puppy face the camera and lick its nose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [0.22146821022033691, 0.4065392017364502], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [0.184454083442688, 0.4065392017364502], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002193.pt"}
{"idx": 2195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hbOXOHEL0tc.mp4", "duration": 116.58324991658326, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment where a woman is checking another woman's pulse on her wrist.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment where a woman is checking another woman's pulse on her wrist.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9485346078872681, -0.862758994102478], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.9485346078872681, -0.862758994102478], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002195.pt"}
{"idx": 2197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hbOXOHEL0tc.mp4", "duration": 116.58324991658326, "prompt_text": "Please find the visual event described by the sentence 'When are the two women doing squats in the gym?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women doing squats in the gym?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 48.0], "base_se_norm": [-0.29663997888565063, -0.17655408382415771], "gt_se_raw": [42.0, 48.0], "gt_se_norm": [-0.2794848680496216, -0.17655408382415771], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002197.pt"}
{"idx": 2199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hbOXOHEL0tc.mp4", "duration": 116.58324991658326, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip showing a close-up of a hand using a computer mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip showing a close-up of a hand using a computer mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.845603883266449, -0.776983380317688], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.862758994102478, -0.7941385507583618], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002199.pt"}
{"idx": 2201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6hbd8N3j7bg.mp4", "duration": 113.65520000000001, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a couple is walking together at night, holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a couple is walking together at night, holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [0.03822791576385498, 0.14381039142608643], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [0.03822791576385498, 0.14381039142608643], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002201.pt"}
{"idx": 2203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6hbd8N3j7bg.mp4", "duration": 113.65520000000001, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the family is having a meal at an outdoor restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the family is having a meal at an outdoor restaurant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 58.0], "base_se_norm": [-0.08495163917541504, 0.020630836486816406], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [-0.08495163917541504, 0.020630836486816406], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002203.pt"}
{"idx": 2205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8rAxqWMKErQ.mp4", "duration": 126.75996666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the child use sign language to communicate with the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child use sign language to communicate with the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 14.0], "base_se_norm": [-0.9842221736907959, -0.7791100740432739], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.968444287776947, -0.9053328633308411], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002205.pt"}
{"idx": 2207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8rAxqWMKErQ.mp4", "duration": 126.75996666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher guide the child's hand to help them spell 'townline'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher guide the child's hand to help them spell 'townline'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 108.0], "base_se_norm": [0.4988960027694702, 0.7040081024169922], "gt_se_raw": [54.0, 59.0], "gt_se_norm": [-0.1479959487915039, -0.06910669803619385], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002207.pt"}
{"idx": 2209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8rAxqWMKErQ.mp4", "duration": 126.75996666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the child spells 'Robert clean' on the device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the child spells 'Robert clean' on the device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 92.0], "base_se_norm": [0.27800607681274414, 0.45156240463256836], "gt_se_raw": [110.0, 122.0], "gt_se_norm": [0.7355637550354004, 0.9248980283737183], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002209.pt"}
{"idx": 2211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1VM2eLhvsSM.mp4", "duration": 59.55949999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a young woman with pigtails and red ribbons sings while holding a Coke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a young woman with pigtails and red ribbons sings while holding a Coke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.4963020086288452, -0.32840269804000854], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.4963020086288452, -0.3619825839996338], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002211.pt"}
{"idx": 2213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1VM2eLhvsSM.mp4", "duration": 59.55949999999999, "prompt_text": "Please find the visual event described by the sentence 'At what time does an Asian woman in a floral kimono sing in profile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does an Asian woman in a floral kimono sing in profile?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.12692350149154663, 0.00739598274230957], "gt_se_raw": [28.0, 31.0], "gt_se_norm": [-0.05976378917694092, 0.04097580909729004], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002213.pt"}
{"idx": 2215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oBDOstVfKQ8.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator first compare the employment rates of Americans with and without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator first compare the employment rates of Americans with and without disabilities?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.030096113681793213], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9030095934867859, -0.4180576801300049], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002215.pt"}
{"idx": 2217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oBDOstVfKQ8.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where the narrator makes a general observation about the employment trend, starting with 'In general'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where the narrator makes a general observation about the employment trend, starting with 'In general'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9030095934867859, -0.030096113681793213], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.22407686710357666, 0.0668942928314209], "start_token_span": [77, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002217.pt"}
{"idx": 2219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kltuUtE6jQo.mp4", "duration": 50.08, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a red laser beam shoots up into the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a red laser beam shoots up into the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 33.0], "base_se_norm": [-0.2012779712677002, 0.31789135932922363], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.2412140965461731, -0.12140578031539917], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002219.pt"}
{"idx": 2221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kltuUtE6jQo.mp4", "duration": 50.08, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the entire logo structure collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the entire logo structure collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 39.0], "base_se_norm": [0.27795517444610596, 0.5575078725814819], "gt_se_raw": [33.0, 39.0], "gt_se_norm": [0.31789135932922363, 0.5575078725814819], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002221.pt"}
{"idx": 2223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9a9XU7D0DIo.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'When can we see an elderly man boxing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see an elderly man boxing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.7111322283744812, -0.6469393968582153], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.7271804809570312, -0.6950840353965759], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002223.pt"}
{"idx": 2225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9a9XU7D0DIo.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a medical professional is measuring an elderly woman's blood pressure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a medical professional is measuring an elderly woman's blood pressure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 12.0], "base_se_norm": [-0.8876625299453735, -0.8074215054512024], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.8716143369674683, -0.8234696984291077], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002225.pt"}
{"idx": 2227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9a9XU7D0DIo.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man is playing with a white dog on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man is playing with a white dog on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 98.0], "base_se_norm": [0.42829060554504395, 0.572724461555481], "gt_se_raw": [89.0, 91.0], "gt_se_norm": [0.42829060554504395, 0.4603869915008545], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002227.pt"}
{"idx": 2229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mMZ6QRf2Uig.mp4", "duration": 168.16666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when flour and soap are stirred together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when flour and soap are stirred together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.3815659284591675, -0.3339940905570984], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [-0.3815659284591675, -0.3221011161804199], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002229.pt"}
{"idx": 2231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mMZ6QRf2Uig.mp4", "duration": 168.16666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where blue glitter is being poured onto the red dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where blue glitter is being poured onto the red dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 125.0], "base_se_norm": [0.45094144344329834, 0.4866204261779785], "gt_se_raw": [123.0, 125.0], "gt_se_norm": [0.4628344774246216, 0.4866204261779785], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002231.pt"}
{"idx": 2233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YTszmB9fqEs.mp4", "duration": 54.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman praise the boy by saying 'good speech'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman praise the boy by saying 'good speech'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 40.0], "base_se_norm": [0.4056720733642578, 0.47965478897094727], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8520345091819763, -0.7780517935752869], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002233.pt"}
{"idx": 2235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YTszmB9fqEs.mp4", "duration": 54.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put her hand in front of her mouth to demonstrate a sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put her hand in front of her mouth to demonstrate a sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [0.22071516513824463, 0.3316892385482788], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.22071516513824463, 0.2946978807449341], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002235.pt"}
{"idx": 2237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YTszmB9fqEs.mp4", "duration": 54.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy says 'I am mad'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy says 'I am mad'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.6300863027572632, -0.519112229347229], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [0.516646146774292, 0.5906288623809814], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002237.pt"}
{"idx": 2239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0b-EatqUvb4.mp4", "duration": 161.11929166666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man with blue hair is rapping behind a wire fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man with blue hair is rapping behind a wire fence.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 35.0], "base_se_norm": [-0.6151919364929199, -0.5655393004417419], "gt_se_raw": [31.0, 32.0], "gt_se_norm": [-0.6151919364929199, -0.6027787923812866], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002239.pt"}
{"idx": 2241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0b-EatqUvb4.mp4", "duration": 161.11929166666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a light blue jacket and grey cap rapping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a light blue jacket and grey cap rapping?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.4786471724510193, -0.4289945363998413], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.4910603165626526, -0.4538208246231079], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002241.pt"}
{"idx": 2243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DexH4oCXw-Y.mp4", "duration": 20.487133333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the boy in the red shirt vocalizing loudly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy in the red shirt vocalizing loudly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 16.0], "base_se_norm": [-0.023777484893798828, 0.5619559288024902], "gt_se_raw": [6.0, 11.0], "gt_se_norm": [-0.41426652669906616, 0.0738447904586792], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002243.pt"}
{"idx": 2245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DexH4oCXw-Y.mp4", "duration": 20.487133333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where an adult voice says 'you have to go through it'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where an adult voice says 'you have to go through it'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.023777484893798828, 0.17146694660186768], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [0.46433377265930176, 0.5619559288024902], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002245.pt"}
{"idx": 2247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DexH4oCXw-Y.mp4", "duration": 20.487133333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the word 'tunnel' is mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the word 'tunnel' is mentioned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.023777484893798828, 0.36671149730682373], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.2690892219543457, 0.46433377265930176], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002247.pt"}
{"idx": 2249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tRAPYopBcc.mp4", "duration": 191.39954166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the little boy is hitting the instrument with sticks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the little boy is hitting the instrument with sticks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 65.0], "base_se_norm": [-0.36258989572525024, -0.3207924962043762], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.36258989572525024, -0.3207924962043762], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002249.pt"}
{"idx": 2251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tRAPYopBcc.mp4", "duration": 191.39954166666666, "prompt_text": "Please find the visual event described by the sentence 'A woman and a little boy are posing for a picture. Pinpoint the time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman and a little boy are posing for a picture. Pinpoint the time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.7596650123596191, -0.7178676128387451], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [-0.7596650123596191, -0.7283170223236084], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002251.pt"}
{"idx": 2253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gu0wmiIngAw.mp4", "duration": 120.45833333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man playing the guitar outdoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man playing the guitar outdoors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.9003804922103882, -0.8007609844207764], "gt_se_raw": [6.0, 11.0], "gt_se_norm": [-0.9003804922103882, -0.8173642158508301], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002253.pt"}
{"idx": 2255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gu0wmiIngAw.mp4", "duration": 120.45833333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man hanging upside down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man hanging upside down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 66.0], "base_se_norm": [0.06260812282562256, 0.09581458568572998], "gt_se_raw": [65.0, 66.0], "gt_se_norm": [0.07921135425567627, 0.09581458568572998], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002255.pt"}
{"idx": 2257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gu0wmiIngAw.mp4", "duration": 120.45833333333333, "prompt_text": "Please find the visual event described by the sentence 'When does a man wearing an oxygen mask and a cape appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man wearing an oxygen mask and a cape appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 61.0], "base_se_norm": [-0.020408213138580322, 0.012798309326171875], "gt_se_raw": [60.0, 61.0], "gt_se_norm": [-0.003804922103881836, 0.012798309326171875], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002257.pt"}
{"idx": 2259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bR2Uon5BWC0.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'When does the bald man grab the other man's hand and start fighting back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bald man grab the other man's hand and start fighting back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 40.0], "base_se_norm": [-0.6311380863189697, -0.5081841349601746], "gt_se_raw": [31.0, 38.0], "gt_se_norm": [-0.6188427209854126, -0.5327749252319336], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002259.pt"}
{"idx": 2261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bR2Uon5BWC0.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'When is the man with the injured hand lying on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man with the injured hand lying on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 111.0], "base_se_norm": [-0.06554985046386719, 0.3647890090942383], "gt_se_raw": [76.0, 84.0], "gt_se_norm": [-0.06554985046386719, 0.032813310623168945], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002261.pt"}
{"idx": 2263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdQwlOy78lg.mp4", "duration": 101.1, "prompt_text": "Please find the visual event described by the sentence 'When does the person demonstrate how to place their finger into the indentation of the pen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person demonstrate how to place their finger into the indentation of the pen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.7823936939239502, -0.6834816932678223], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.7823936939239502, -0.7032641172409058], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002263.pt"}
{"idx": 2265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdQwlOy78lg.mp4", "duration": 101.1, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the back of the pen's packaging is displayed to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the back of the pen's packaging is displayed to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 53.0], "base_se_norm": [-0.24826902151107788, 0.04846692085266113], "gt_se_raw": [38.0, 54.0], "gt_se_norm": [-0.24826902151107788, 0.06824922561645508], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002265.pt"}
{"idx": 2267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdQwlOy78lg.mp4", "duration": 101.1, "prompt_text": "Please find the visual event described by the sentence 'At what point is the person clicking the pen clip as if it were a fidget toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the person clicking the pen clip as if it were a fidget toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 99.0], "base_se_norm": [0.8595449924468994, 0.9584569931030273], "gt_se_raw": [94.0, 99.0], "gt_se_norm": [0.8595449924468994, 0.9584569931030273], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002267.pt"}
{"idx": 2269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4w7sVSMbjyM.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two men being interviewed in front of a fireplace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two men being interviewed in front of a fireplace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 46.0], "base_se_norm": [-0.6524908542633057, -0.5155932903289795], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [-0.6524908542633057, -0.610368549823761], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002269.pt"}
{"idx": 2271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4w7sVSMbjyM.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man is cross-country snowboarding on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man is cross-country snowboarding on a treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.21101665496826172, 0.2426084280014038], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [0.21101665496826172, 0.2426084280014038], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002271.pt"}
{"idx": 2273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w38j0ea8HO4.mp4", "duration": 126.86006666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in red stab the man's hand with a fork?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in red stab the man's hand with a fork?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [-0.1329028606414795, -0.08560669422149658], "gt_se_raw": [59.0, 61.0], "gt_se_norm": [-0.06984126567840576, -0.038310468196868896], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002273.pt"}
{"idx": 2275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w38j0ea8HO4.mp4", "duration": 126.86006666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman wearing a red outfit and a large hat make her entrance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman wearing a red outfit and a large hat make her entrance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 35.0], "base_se_norm": [-0.6373957395553589, -0.4482109546661377], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.6216303110122681, -0.4955071210861206], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002275.pt"}
{"idx": 2277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w38j0ea8HO4.mp4", "duration": 126.86006666666667, "prompt_text": "Please find the visual event described by the sentence 'When does a man order ten chocolate sundaes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man order ten chocolate sundaes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 122.0], "base_se_norm": [0.8760828971862793, 0.9233790636062622], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.9684692025184631, -0.9369384050369263], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002277.pt"}
{"idx": 2279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggqrEMtA3TU.mp4", "duration": 377.28, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket pour sugar into a coffee cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket pour sugar into a coffee cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 106.0], "base_se_norm": [-0.45398640632629395, -0.43808311223983765], "gt_se_raw": [103.0, 106.0], "gt_se_norm": [-0.45398640632629395, -0.43808311223983765], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002279.pt"}
{"idx": 2281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggqrEMtA3TU.mp4", "duration": 377.28, "prompt_text": "Please find the visual event described by the sentence 'When is coffee discussed as a way to 'dream faster'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 175.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is coffee discussed as a way to 'dream faster'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 175.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 175.0], "base_se_norm": [-0.13592028617858887, -0.0723070502281189], "gt_se_raw": [160.0, 176.0], "gt_se_norm": [-0.15182358026504517, -0.0670059323310852], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002281.pt"}
{"idx": 2283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tw7PR89IOI.mp4", "duration": 40.073366666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the boy open the mailbox with assistance from the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy open the mailbox with assistance from the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 38.0], "base_se_norm": [0.347528338432312, 0.8965214490890503], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [0.4473452568054199, 0.6469790935516357], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002283.pt"}
{"idx": 2285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tw7PR89IOI.mp4", "duration": 40.073366666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman pushing the boy in the wheelchair past the green bushes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman pushing the boy in the wheelchair past the green bushes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 19.0], "base_se_norm": [-0.8003661632537842, -0.051739275455474854], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.2513731122016907, -0.0018308162689208984], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002285.pt"}
{"idx": 2287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_tw7PR89IOI.mp4", "duration": 40.073366666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the beginning of the video where the woman starts pushing the wheelchair on the driveway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the beginning of the video where the woman starts pushing the wheelchair on the driveway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.950091540813446, -0.5009154081344604], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.950091540813446, -0.7005492448806763], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002287.pt"}
{"idx": 2289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGAVcg_xUJE.mp4", "duration": 131.56476666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man shouts loudly with his mouth wide open and hands raised.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man shouts loudly with his mouth wide open and hands raised.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 75.0], "base_se_norm": [0.09451794624328613, 0.14012277126312256], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.7415721416473389, -0.6959672570228577], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002289.pt"}
{"idx": 2291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGAVcg_xUJE.mp4", "duration": 131.56476666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the word 'NATO' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the word 'NATO' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 49.0], "base_se_norm": [-0.34632956981658936, -0.2551197409629822], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.34632956981658936, -0.2703213691711426], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002291.pt"}
{"idx": 2293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3OhB-y7Tls8.mp4", "duration": 132.2146064760104, "prompt_text": "Please find the visual event described by the sentence 'When can we see the cable car in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the cable car in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [-0.2134000062942505, -0.16801923513412476], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.2134000062942505, -0.1831461787223816], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002293.pt"}
{"idx": 2295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3OhB-y7Tls8.mp4", "duration": 132.2146064760104, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a path with steps is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a path with steps is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 70.0], "base_se_norm": [-0.09238463640213013, 0.058884620666503906], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [-0.07725775241851807, -0.047003865242004395], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002295.pt"}
{"idx": 2297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3OhB-y7Tls8.mp4", "duration": 132.2146064760104, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a person in a wheelchair is being pushed up a ramp by an assistant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a person in a wheelchair is being pushed up a ramp by an assistant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 110.0], "base_se_norm": [0.61858069896698, 0.6639615297317505], "gt_se_raw": [108.0, 110.0], "gt_se_norm": [0.6337076425552368, 0.6639615297317505], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002297.pt"}
{"idx": 2299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HvPbH30KWLE.mp4", "duration": 119.61949999999999, "prompt_text": "Please find the visual event described by the sentence 'At what point are the men dancing inside a vehicle with purple and green disco lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the men dancing inside a vehicle with purple and green disco lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 51.0], "base_se_norm": [-0.5820079445838928, -0.14729619026184082], "gt_se_raw": [25.0, 32.0], "gt_se_norm": [-0.5820079445838928, -0.4649701714515686], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002299.pt"}
{"idx": 2301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HvPbH30KWLE.mp4", "duration": 119.61949999999999, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a person is holding up a phone with a pineapple case to take a picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a person is holding up a phone with a pineapple case to take a picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 66.0], "base_se_norm": [0.07005965709686279, 0.10349905490875244], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [0.07005965709686279, 0.10349905490875244], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002301.pt"}
{"idx": 2303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQD6gJMnWVs.mp4", "duration": 112.71260000000001, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a jean jacket speaking in a supermarket aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a jean jacket speaking in a supermarket aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 13.0], "base_se_norm": [-0.964511513710022, -0.769324779510498], "gt_se_raw": [3.0, 13.0], "gt_se_norm": [-0.946767270565033, -0.769324779510498], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002303.pt"}
{"idx": 2305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQD6gJMnWVs.mp4", "duration": 112.71260000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a green polo shirt put groceries into a plastic bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a green polo shirt put groceries into a plastic bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [-0.07729923725128174, -0.024066507816314697], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [-0.07729923725128174, -0.024066507816314697], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002305.pt"}
{"idx": 2307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQD6gJMnWVs.mp4", "duration": 112.71260000000001, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that shows a screen recording of the Woolworths online shopping website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that shows a screen recording of the Woolworths online shopping website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 83.0], "base_se_norm": [0.2598414421081543, 0.47277235984802246], "gt_se_raw": [71.0, 83.0], "gt_se_norm": [0.2598414421081543, 0.47277235984802246], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002307.pt"}
{"idx": 2309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yqFxYu15J6E.mp4", "duration": 35.618916666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl open the red binder to look at the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl open the red binder to look at the calendar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.49465054273605347, -0.2700507640838623], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.49465054273605347, -0.3262007236480713], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002309.pt"}
{"idx": 2311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yqFxYu15J6E.mp4", "duration": 35.618916666666664, "prompt_text": "Please find the visual event described by the sentence 'When is the girl choosing a sticker from the orange folder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl choosing a sticker from the orange folder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [0.12299883365631104, 0.4598984718322754], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [0.17914879322052002, 0.4598984718322754], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002311.pt"}
{"idx": 2313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CecE0asmNwc.mp4", "duration": 188.39654166666665, "prompt_text": "Please find the visual event described by the sentence 'When does a worker install a new battery into a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a worker install a new battery into a wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 47.0], "base_se_norm": [-0.5541319847106934, -0.5010523796081543], "gt_se_raw": [42.0, 47.0], "gt_se_norm": [-0.5541319847106934, -0.5010523796081543], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002313.pt"}
{"idx": 2315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CecE0asmNwc.mp4", "duration": 188.39654166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man putting a folded wheelchair into the back of a truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man putting a folded wheelchair into the back of a truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.4798206090927124, -0.43735700845718384], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [-0.46920472383499146, -0.43735700845718384], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002315.pt"}
{"idx": 2317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CecE0asmNwc.mp4", "duration": 188.39654166666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see a young person smiling while sitting in a motorized wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a young person smiling while sitting in a motorized wheelchair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 112.0], "base_se_norm": [0.08282244205474854, 0.18898141384124756], "gt_se_raw": [112.0, 115.0], "gt_se_norm": [0.18898141384124756, 0.22082912921905518], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002317.pt"}
{"idx": 2319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CeN6PMEOQEA.mp4", "duration": 147.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man presents the bottle of lingonberry drink to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man presents the bottle of lingonberry drink to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.8096317052841187, -0.755240797996521], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.8096317052841187, -0.755240797996521], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002319.pt"}
{"idx": 2321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CeN6PMEOQEA.mp4", "duration": 147.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the man chugs the entire glass of the red drink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the man chugs the entire glass of the red drink.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 130.0], "base_se_norm": [0.4277620315551758, 0.7677054405212402], "gt_se_raw": [105.0, 131.0], "gt_se_norm": [0.4277620315551758, 0.7813031673431396], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002321.pt"}
{"idx": 2323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u3VVUu-lZsM.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When is the chart showing employment statistics for disabled and non-disabled Americans visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the chart showing employment statistics for disabled and non-disabled Americans visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002323.pt"}
{"idx": 2325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u3VVUu-lZsM.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the period where the data for the year 1991 is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the period where the data for the year 1991 is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002325.pt"}
{"idx": 2327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u3VVUu-lZsM.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the employment percentage for the disabled in 2010 is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the employment percentage for the disabled in 2010 is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002327.pt"}
{"idx": 2329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eA-GJNebJj0.mp4", "duration": 69.70296666666667, "prompt_text": "Please find the visual event described by the sentence 'A hand pumps water from a well.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A hand pumps water from a well.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 23.0], "base_se_norm": [-0.5982954502105713, -0.34005677700042725], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.6269886493682861, -0.4835227131843567], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002329.pt"}
{"idx": 2331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eA-GJNebJj0.mp4", "duration": 69.70296666666667, "prompt_text": "Please find the visual event described by the sentence 'Show a close-up of the wine bottles in the box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show a close-up of the wine bottles in the box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 51.0], "base_se_norm": [0.20511376857757568, 0.4633524417877197], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [0.23380696773529053, 0.34857964515686035], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002331.pt"}
{"idx": 2333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6nX0lYo0Uno.mp4", "duration": 98.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man introduce himself by stating his full name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man introduce himself by stating his full name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9190556406974792], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.9392917156219482, -0.9190556406974792], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002333.pt"}
{"idx": 2335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6nX0lYo0Uno.mp4", "duration": 98.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does he talk about working at a local financial firm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does he talk about working at a local financial firm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 26.0], "base_se_norm": [-0.6559865474700928, -0.4738617539405823], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.5750421285629272, -0.5143339037895203], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002335.pt"}
{"idx": 2337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6nX0lYo0Uno.mp4", "duration": 98.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment he discusses the global pandemic breaking out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment he discusses the global pandemic breaking out.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 16.0], "base_se_norm": [-0.979763925075531, -0.676222562789917], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [0.11298477649688721, 0.1534569263458252], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002337.pt"}
{"idx": 2339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGMab_ZhQTM.mp4", "duration": 134.63040304234153, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man point with both of his index fingers while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man point with both of his index fingers while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.5543354749679565, -0.5097689628601074], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.5543354749679565, -0.5246244668960571], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002339.pt"}
{"idx": 2341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hGMab_ZhQTM.mp4", "duration": 134.63040304234153, "prompt_text": "Please find the visual event described by the sentence 'When does the man describe the 'Braille 'n Speak' device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man describe the 'Braille 'n Speak' device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 76.0], "base_se_norm": [-0.8514451384544373, 0.12901687622070312], "gt_se_raw": [64.0, 85.0], "gt_se_norm": [-0.04924893379211426, 0.26271629333496094], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002341.pt"}
{"idx": 2343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1WsDtn-feuI.mp4", "duration": 13.066666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man walk towards the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man walk towards the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.5408163070678711], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.6938775777816772], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002343.pt"}
{"idx": 2345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1WsDtn-feuI.mp4", "duration": 13.066666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man, hanging from the ceiling, try to touch the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man, hanging from the ceiling, try to touch the pie?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.3877550959587097, 0.22448980808258057], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [0.3775510787963867, 0.5306122303009033], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002345.pt"}
{"idx": 2347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1WsDtn-feuI.mp4", "duration": 13.066666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man hangs upside down above an empty plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man hangs upside down above an empty plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [0.22448980808258057, 0.8367346525192261], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.6836735010147095, 0.9897959232330322], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002347.pt"}
{"idx": 2349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NAizsUKpafs.mp4", "duration": 129.5294, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker explain the full travel itinerary, including driving and staying in a hotel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker explain the full travel itinerary, including driving and staying in a hotel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 48.0], "base_se_norm": [-0.6911897659301758, -0.25885552167892456], "gt_se_raw": [66.0, 75.0], "gt_se_norm": [0.01907360553741455, 0.15803825855255127], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002349.pt"}
{"idx": 2351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NAizsUKpafs.mp4", "duration": 129.5294, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker compare Jake to her other children, stating they 'just got it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker compare Jake to her other children, stating they 'just got it'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [-0.16621249914169312, -0.0735694169998169], "gt_se_raw": [92.0, 94.0], "gt_se_norm": [0.4205268621444702, 0.45140790939331055], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002351.pt"}
{"idx": 2353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTTxroRR-G0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a teacher is reading to a group of children on a colorful butterfly rug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a teacher is reading to a group of children on a colorful butterfly rug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002353.pt"}
{"idx": 2355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTTxroRR-G0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a teacher in a red shirt shows a book with a picture of a cow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a teacher in a red shirt shows a book with a picture of a cow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002355.pt"}
{"idx": 2357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTTxroRR-G0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what time is the text 'LENGTH of PROGRAM' visible on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the text 'LENGTH of PROGRAM' visible on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [0.049278974533081055, 0.44275856018066406], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002357.pt"}
{"idx": 2359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/huT5__BqY_U.mp4", "duration": 166.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the cows start rushing out of the pen and down the path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the cows start rushing out of the pen and down the path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 36.0], "base_se_norm": [-0.7236683964729309, -0.5674809217453003], "gt_se_raw": [24.0, 37.0], "gt_se_norm": [-0.7116539478302002, -0.5554665327072144], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002359.pt"}
{"idx": 2361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/huT5__BqY_U.mp4", "duration": 166.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the cows begin to calm down and start grazing on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the cows begin to calm down and start grazing on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 139.0], "base_se_norm": [0.20144176483154297, 0.67000412940979], "gt_se_raw": [76.0, 100.0], "gt_se_norm": [-0.08690422773361206, 0.20144176483154297], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002361.pt"}
{"idx": 2363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQ7YxcW9Vyk.mp4", "duration": 91.82506666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the boy celebrating his 5th birthday with a cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy celebrating his 5th birthday with a cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 56.0], "base_se_norm": [0.0890272855758667, 0.2197105884552002], "gt_se_raw": [50.0, 56.0], "gt_se_norm": [0.0890272855758667, 0.2197105884552002], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002363.pt"}
{"idx": 2365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQ7YxcW9Vyk.mp4", "duration": 91.82506666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a newborn baby is lying in a hospital bed with a red cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a newborn baby is lying in a hospital bed with a red cap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 17.0], "base_se_norm": [-0.8257556557655334, -0.6297307014465332], "gt_se_raw": [7.0, 17.0], "gt_se_norm": [-0.8475362062454224, -0.6297307014465332], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002365.pt"}
{"idx": 2367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pQ7YxcW9Vyk.mp4", "duration": 91.82506666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a toddler in a blue sweater sticking his tongue out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a toddler in a blue sweater sticking his tongue out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 69.0], "base_se_norm": [0.32861328125, 0.5028575658798218], "gt_se_raw": [61.0, 69.0], "gt_se_norm": [0.32861328125, 0.5028575658798218], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002367.pt"}
{"idx": 2369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s_RV4Btuv2c.mp4", "duration": 36.3363, "prompt_text": "Please find the visual event described by the sentence 'When is the man filling the pot with water from the tap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man filling the pot with water from the tap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 23.0], "base_se_norm": [-0.1743793487548828, 0.26595163345336914], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.06429660320281982, 0.26595163345336914], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002369.pt"}
{"idx": 2371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s_RV4Btuv2c.mp4", "duration": 36.3363, "prompt_text": "Please find the visual event described by the sentence 'A person in a hooded jacket gets on a bus and sits down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A person in a hooded jacket gets on a bus and sits down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [0.7613240480422974, 0.9264482259750366], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [0.7613240480422974, 0.8714067935943604], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002371.pt"}
{"idx": 2373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SBTL1vI4-mc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman with reddish hair using CAD software?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman with reddish hair using CAD software?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.540437638759613], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002373.pt"}
{"idx": 2375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SBTL1vI4-mc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a young man with dark hair is operating a large format printer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a young man with dark hair is operating a large format printer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.015223503112792969, 0.3786870241165161], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002375.pt"}
{"idx": 2377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SBTL1vI4-mc.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'LENGTH of PROGRAM' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'LENGTH of PROGRAM' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.05042827129364014, 0.3786870241165161], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002377.pt"}
{"idx": 2379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YvcTBNSuiGk.mp4", "duration": 231.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the woman in the blue shirt points to the paper to guide the boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the woman in the blue shirt points to the paper to guide the boy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 72.0], "base_se_norm": [-0.4132029414176941, -0.37868547439575195], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.40457355976104736, -0.3873147964477539], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002379.pt"}
{"idx": 2381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YvcTBNSuiGk.mp4", "duration": 231.76666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the boy ask if the item on his card is something you can wear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy ask if the item on his card is something you can wear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.9137063026428223, -0.8878182172775269], "gt_se_raw": [147.0, 149.0], "gt_se_norm": [0.26851725578308105, 0.28577589988708496], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002381.pt"}
{"idx": 2383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_NWYeVyZz9I.mp4", "duration": 176.09258333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the moose character playing a trumpet by the lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the moose character playing a trumpet by the lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.9545693397521973, -0.8977810740470886], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.9545693397521973, -0.9091386795043945], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002383.pt"}
{"idx": 2385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_NWYeVyZz9I.mp4", "duration": 176.09258333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman in the plaid shirt adds purple sparkles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman in the plaid shirt adds purple sparkles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.3866862654685974, -0.35261332988739014], "gt_se_raw": [55.0, 56.0], "gt_se_norm": [-0.37532860040664673, -0.3639709949493408], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002385.pt"}
{"idx": 2387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_NWYeVyZz9I.mp4", "duration": 176.09258333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the character in the gorilla suit add purple bananas to the stew?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character in the gorilla suit add purple bananas to the stew?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 107.0], "base_se_norm": [0.1357661485671997, 0.2152698040008545], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.2152698040008545, 0.23798513412475586], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002387.pt"}
{"idx": 2389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fSVp2m1ahwk.mp4", "duration": 29.56, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where a group of people are sitting in a circle for a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where a group of people are sitting in a circle for a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [-0.5263869762420654, -0.3910689949989319], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.5940459966659546, -0.45872801542282104], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002389.pt"}
{"idx": 2391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fSVp2m1ahwk.mp4", "duration": 29.56, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a man with glasses and a grey t-shirt is speaking to a group.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a man with glasses and a grey t-shirt is speaking to a group.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [0.15020298957824707, 0.35318005084991455], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.21786201000213623, 0.35318005084991455], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002391.pt"}
{"idx": 2393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-8Th1fNuXA.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when a man in a grey hoodie is using a multimeter on the motor control system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when a man in a grey hoodie is using a multimeter on the motor control system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002393.pt"}
{"idx": 2395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-8Th1fNuXA.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the close-up shot where a hand is adjusting a small mechanical component.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the close-up shot where a hand is adjusting a small mechanical component.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [0.18043887615203857, 0.3771786689758301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002395.pt"}
{"idx": 2397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-8Th1fNuXA.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where 'LENGTH of PROGRAM' is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where 'LENGTH of PROGRAM' is displayed on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.049278974533081055, 0.3771786689758301], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002397.pt"}
{"idx": 2399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsEOupmYmF0.mp4", "duration": 156.22273333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is there a close-up shot of the woman placing an olive in her mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there a close-up shot of the woman placing an olive in her mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 41.0], "base_se_norm": [-0.5135151147842407, -0.47510838508605957], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.5135151147842407, -0.47510838508605957], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002399.pt"}
{"idx": 2401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsEOupmYmF0.mp4", "duration": 156.22273333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put olives on her fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put olives on her fingers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [-0.014227986335754395, 0.03698098659515381], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.014227986335754395, 0.03698098659515381], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002401.pt"}
{"idx": 2403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cEfD-Mr7m2c.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When is the employment rate for disabled Americans in 1991 displayed on the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the employment rate for disabled Americans in 1991 displayed on the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.5150480270385742], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002403.pt"}
{"idx": 2405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cEfD-Mr7m2c.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator mention that the employment rate for non-disabled people is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator mention that the employment rate for non-disabled people is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9030095934867859, -0.4180576801300049], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9030095934867859, -0.4180576801300049], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002405.pt"}
{"idx": 2407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cEfD-Mr7m2c.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker say people with disabilities are employed at nearly half the rate of those without?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker say people with disabilities are employed at nearly half the rate of those without?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.12708652019500732, 0.35786545276641846], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002407.pt"}
{"idx": 2409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nqkyzpaoMug.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When is a person receiving a facial treatment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person receiving a facial treatment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.4097805619239807, -0.08188092708587646], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002409.pt"}
{"idx": 2411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nqkyzpaoMug.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where two women are looking at a book together in a hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where two women are looking at a book together in a hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002411.pt"}
{"idx": 2413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KTtkpl7T9sg.mp4", "duration": 118.70191666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman walk into the scene from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman walk into the scene from the right?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.8989064693450928, -0.8315107226371765], "gt_se_raw": [6.0, 10.0], "gt_se_norm": [-0.8989064693450928, -0.8315107226371765], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002413.pt"}
{"idx": 2415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KTtkpl7T9sg.mp4", "duration": 118.70191666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the woman realize her phone is dead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the woman realize her phone is dead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 83.0], "base_se_norm": [0.21312272548675537, 0.3984609842300415], "gt_se_raw": [80.0, 83.0], "gt_se_norm": [0.3479142189025879, 0.3984609842300415], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002415.pt"}
{"idx": 2417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KTtkpl7T9sg.mp4", "duration": 118.70191666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a figure appears in the frame on the left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a figure appears in the frame on the left.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 95.0], "base_se_norm": [0.5501012802124023, 0.600648045539856], "gt_se_raw": [92.0, 94.0], "gt_se_norm": [0.5501012802124023, 0.5837991237640381], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002417.pt"}
{"idx": 2419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUenwQAtW8M.mp4", "duration": 84.65123333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time the golden retriever barks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time the golden retriever barks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [0.1576913595199585, 0.20494413375854492], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.08681237697601318, 0.13406503200531006], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002419.pt"}
{"idx": 2421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUenwQAtW8M.mp4", "duration": 84.65123333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the red shirt walk down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the red shirt walk down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.4884603023529053, 0.6065921783447266], "gt_se_raw": [65.0, 71.0], "gt_se_norm": [0.5357130765914917, 0.6774711608886719], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002421.pt"}
{"idx": 2423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-mJqm_ijE6s.mp4", "duration": 82.61586666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where people are having a snowball fight outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where people are having a snowball fight outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9273747205734253, -0.7821241617202759], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.9031662940979004, -0.8305410146713257], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002423.pt"}
{"idx": 2425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-mJqm_ijE6s.mp4", "duration": 82.61586666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman hugs a boy wearing glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman hugs a boy wearing glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 56.0], "base_se_norm": [0.2830464839935303, 0.355671763420105], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [0.2830464839935303, 0.3314633369445801], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002425.pt"}
{"idx": 2427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-mJqm_ijE6s.mp4", "duration": 82.61586666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man with long dark hair raise his glass and say 'Harry'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man with long dark hair raise his glass and say 'Harry'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 77.0], "base_se_norm": [0.8156318664550781, 0.8640487194061279], "gt_se_raw": [75.0, 76.0], "gt_se_norm": [0.8156318664550781, 0.839840292930603], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002427.pt"}
{"idx": 2429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rCZ3SN65kIs.mp4", "duration": 272.98104166666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the Orc commander order his troops to form ranks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the Orc commander order his troops to form ranks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [-0.5017969608306885, -0.4651643633842468], "gt_se_raw": [69.0, 73.0], "gt_se_norm": [-0.49447041749954224, -0.4651643633842468], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002429.pt"}
{"idx": 2431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rCZ3SN65kIs.mp4", "duration": 272.98104166666667, "prompt_text": "Please find the visual event described by the sentence 'When do the Rohirrim blow their horns to signal the charge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the Rohirrim blow their horns to signal the charge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 173.0], "base_se_norm": [0.2381812334060669, 0.2674872875213623], "gt_se_raw": [170.0, 173.0], "gt_se_norm": [0.24550771713256836, 0.2674872875213623], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002431.pt"}
{"idx": 2433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aHi8idsTvVM.mp4", "duration": 96.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see Miranda requesting Gail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see Miranda requesting Gail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.8760757446289062], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.8760757446289062], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002433.pt"}
{"idx": 2435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aHi8idsTvVM.mp4", "duration": 96.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where Miranda is placing an order for a coffee at the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where Miranda is placing an order for a coffee at the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 53.0], "base_se_norm": [-0.2564544081687927, 0.09466433525085449], "gt_se_raw": [36.0, 44.0], "gt_se_norm": [-0.2564544081687927, -0.09122204780578613], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002435.pt"}
{"idx": 2437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aHi8idsTvVM.mp4", "duration": 96.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where Miranda delivers an order with the help of a woman in a red shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where Miranda delivers an order with the help of a woman in a red shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 96.0], "base_se_norm": [0.6110154390335083, 0.9827882051467896], "gt_se_raw": [78.0, 94.0], "gt_se_norm": [0.6110154390335083, 0.9414801597595215], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002437.pt"}
{"idx": 2439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oKficmlxzaI.mp4", "duration": 190.4903, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a fledgling is standing by the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a fledgling is standing by the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 39.0], "base_se_norm": [-0.6430264711380005, -0.5905303359031677], "gt_se_raw": [34.0, 39.0], "gt_se_norm": [-0.6430264711380005, -0.5905303359031677], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002439.pt"}
{"idx": 2441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oKficmlxzaI.mp4", "duration": 190.4903, "prompt_text": "Please find the visual event described by the sentence 'At what point does one eagle try to steal a fish from another eagle in mid-air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one eagle try to steal a fish from another eagle in mid-air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 133.0], "base_se_norm": [0.1549144983291626, 0.3963966369628906], "gt_se_raw": [111.0, 117.0], "gt_se_norm": [0.1654137372970581, 0.2284090518951416], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002441.pt"}
{"idx": 2443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Grz6lB5YX9I.mp4", "duration": 22.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the striped shirt clap her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the striped shirt clap her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.7293232679367065, -0.548872172832489], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7293232679367065, -0.45864659547805786], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002443.pt"}
{"idx": 2445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Grz6lB5YX9I.mp4", "duration": 22.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the baby start crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the baby start crying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [0.17293238639831543, 0.6240601539611816], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [0.17293238639831543, 0.4436091184616089], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002445.pt"}
{"idx": 2447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Grz6lB5YX9I.mp4", "duration": 22.166666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman lowers her head toward the baby's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman lowers her head toward the baby's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.27819544076919556, -0.007518768310546875], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.18796992301940918, -0.007518768310546875], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002447.pt"}
{"idx": 2449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JOCtdw9FG-s.mp4", "duration": 83.79204166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the chef is wrapping the tiny burrito.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the chef is wrapping the tiny burrito.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.6658393740653992, -0.5464962720870972], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.6419707536697388, -0.5703648924827576], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002449.pt"}
{"idx": 2451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JOCtdw9FG-s.mp4", "duration": 83.79204166666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the hamster is first fed the burrito.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the hamster is first fed the burrito.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 45.0], "base_se_norm": [-0.21233564615249634, 0.07408773899078369], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.21233564615249634, -0.14072978496551514], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002451.pt"}
{"idx": 2453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Osw1_0jMM4M.mp4", "duration": 127.36056666666667, "prompt_text": "Please find the visual event described by the sentence 'When are hands shown forming a peace sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are hands shown forming a peace sign?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.8586689829826355, -0.7801517248153687], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8586689829826355, -0.7958551645278931], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002453.pt"}
{"idx": 2455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Osw1_0jMM4M.mp4", "duration": 127.36056666666667, "prompt_text": "Please find the visual event described by the sentence 'At what moment does a girl look at a globe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does a girl look at a globe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 67.0], "base_se_norm": [0.005020618438720703, 0.052131056785583496], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [0.020724058151245117, 0.052131056785583496], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002455.pt"}
{"idx": 2457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Osw1_0jMM4M.mp4", "duration": 127.36056666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a diverse group of young people posing together for a photo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a diverse group of young people posing together for a photo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 61.0], "base_se_norm": [-0.08920001983642578, -0.04208970069885254], "gt_se_raw": [58.0, 61.0], "gt_se_norm": [-0.08920001983642578, -0.04208970069885254], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002457.pt"}
{"idx": 2459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzWq3LqxAIc.mp4", "duration": 23.857166666666668, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the person pours the food into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the person pours the food into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.07784521579742432, 0.2574838399887085], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [0.005987048149108887, 0.2574838399887085], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002459.pt"}
{"idx": 2461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qzWq3LqxAIc.mp4", "duration": 23.857166666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point is the person closing the lid of the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the person closing the lid of the container?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [0.5089805126190186, 0.7604773044586182], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [0.5089805126190186, 0.7604773044586182], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002461.pt"}
{"idx": 2463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S0zfR9DTwSY.mp4", "duration": 15.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the children making waves with the colorful parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the children making waves with the colorful parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.9027484655380249], "gt_se_raw": [0.0, 15.0], "gt_se_norm": [-1.0, 0.9027484655380249], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002463.pt"}
{"idx": 2465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S0zfR9DTwSY.mp4", "duration": 15.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl with long blonde hair on the right try to touch the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl with long blonde hair on the right try to touch the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [0.014799118041992188, 0.2684990167617798], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [0.1416490077972412, 0.3953489065170288], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002465.pt"}
{"idx": 2467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S0zfR9DTwSY.mp4", "duration": 15.766666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the beach ball bounce highest near the center of the parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the beach ball bounce highest near the center of the parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.23890060186386108, 0.014799118041992188], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.6194503307342529, -0.3657504916191101], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002467.pt"}
{"idx": 2469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3cOSt2QVJf4.mp4", "duration": 123.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a pig and a sheep are twirling together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a pig and a sheep are twirling together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 41.0], "base_se_norm": [-0.5463137626647949, -0.33567380905151367], "gt_se_raw": [28.0, 40.0], "gt_se_norm": [-0.5463137626647949, -0.3518768548965454], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002469.pt"}
{"idx": 2471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3cOSt2QVJf4.mp4", "duration": 123.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the little chicks are scrambling and saying 'CHEEP'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the little chicks are scrambling and saying 'CHEEP'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 82.0], "base_se_norm": [0.16662168502807617, 0.32865238189697266], "gt_se_raw": [72.0, 82.0], "gt_se_norm": [0.16662168502807617, 0.32865238189697266], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002471.pt"}
{"idx": 2473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymPpIzaanhY.mp4", "duration": 5.633333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man shouting with his mouth wide open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man shouting with his mouth wide open?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.28994083404541016], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.6449704170227051], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002473.pt"}
{"idx": 2475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymPpIzaanhY.mp4", "duration": 5.633333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man is wearing a large bow tie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man is wearing a large bow tie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7751479148864746], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7751479148864746], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002475.pt"}
{"idx": 2477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymPpIzaanhY.mp4", "duration": 5.633333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man emphatically say the word 'nothing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man emphatically say the word 'nothing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [0.06508874893188477, 0.7751479148864746], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.6449704170227051, -0.28994083404541016], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002477.pt"}
{"idx": 2479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qoh3YkxuwVo.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'At what point does the young man state that he has quit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the young man state that he has quit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 87.0], "base_se_norm": [0.02683699131011963, 0.07632303237915039], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.7649409770965576, -0.7401978969573975], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002479.pt"}
{"idx": 2481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qoh3YkxuwVo.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'When does the older man mention that he was a player on the team?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the older man mention that he was a player on the team?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 97.0], "base_se_norm": [-0.010277628898620605, 0.20003831386566162], "gt_se_raw": [99.0, 101.0], "gt_se_norm": [0.22478139400482178, 0.24952447414398193], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002481.pt"}
{"idx": 2483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vcrOYQbg3hY.mp4", "duration": 115.40695833333332, "prompt_text": "Please find the visual event described by the sentence 'When does Dr. Kim LeDuff shake hands with a female student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Dr. Kim LeDuff shake hands with a female student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.6880604028701782, -0.6534004211425781], "gt_se_raw": [19.0, 20.0], "gt_se_norm": [-0.670730471611023, -0.6534004211425781], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002483.pt"}
{"idx": 2485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vcrOYQbg3hY.mp4", "duration": 115.40695833333332, "prompt_text": "Please find the visual event described by the sentence 'At what moment do we see a group of people walking out of a building's entrance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment do we see a group of people walking out of a building's entrance?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.22015100717544556, -0.09884119033813477], "gt_se_raw": [45.0, 50.0], "gt_se_norm": [-0.22015100717544556, -0.1335011124610901], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002485.pt"}
{"idx": 2487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vcrOYQbg3hY.mp4", "duration": 115.40695833333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when three people are having a meeting at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when three people are having a meeting at a round table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 112.0], "base_se_norm": [0.8369776010513306, 0.9409574270248413], "gt_se_raw": [106.0, 111.0], "gt_se_norm": [0.8369776010513306, 0.923627495765686], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002487.pt"}
{"idx": 2489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oiX7oBoavlA.mp4", "duration": 154.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does a character write 'Wolf' in a notebook, referencing the 'Death Note' anime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a character write 'Wolf' in a notebook, referencing the 'Death Note' anime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [-0.3258426785469055, -0.26101988554000854], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [-0.3258426785469055, -0.26101988554000854], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002489.pt"}
{"idx": 2491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oiX7oBoavlA.mp4", "duration": 154.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a character is defeated, and the scene transitions to a clip of Majin Buu from Dragon Ball Z.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a character is defeated, and the scene transitions to a clip of Majin Buu from Dragon Ball Z.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 134.0], "base_se_norm": [0.6983578205108643, 0.7372515201568604], "gt_se_raw": [131.0, 134.0], "gt_se_norm": [0.6983578205108643, 0.7372515201568604], "start_token_span": [80, 85], "end_token_span": [87, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002491.pt"}
{"idx": 2493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TGrF7g4Z8EE.mp4", "duration": 135.59379166666665, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the ladybug on the railing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the ladybug on the railing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.660751461982727, -0.6312515735626221], "gt_se_raw": [23.0, 24.0], "gt_se_norm": [-0.660751461982727, -0.6460014581680298], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002493.pt"}
{"idx": 2495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TGrF7g4Z8EE.mp4", "duration": 135.59379166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the presenter lists the materials needed for the home experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the presenter lists the materials needed for the home experiment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.20350337028503418, -0.07075393199920654], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [-0.1740034818649292, -0.11500376462936401], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002495.pt"}
{"idx": 2497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfTfAIe9y84.mp4", "duration": 159.52603333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker first introduce the film 'The Theory of Everything'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker first introduce the film 'The Theory of Everything'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, -0.8244800567626953], "gt_se_raw": [4.0, 14.0], "gt_se_norm": [-0.9498514533042908, -0.8244800567626953], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002497.pt"}
{"idx": 2499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfTfAIe9y84.mp4", "duration": 159.52603333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is using a computer program on his wheelchair to communicate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is using a computer program on his wheelchair to communicate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 42.0], "base_se_norm": [-0.5612001419067383, -0.47344017028808594], "gt_se_raw": [35.0, 42.0], "gt_se_norm": [-0.5612001419067383, -0.47344017028808594], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002499.pt"}
{"idx": 2501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PfTfAIe9y84.mp4", "duration": 159.52603333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man, now older and in a wheelchair, is giving a speech to a large audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man, now older and in a wheelchair, is giving a speech to a large audience.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 144.0], "base_se_norm": [0.6172908544540405, 0.8053480386734009], "gt_se_raw": [135.0, 144.0], "gt_se_norm": [0.6925137042999268, 0.8053480386734009], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002501.pt"}
{"idx": 2503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E_Ci-pAL4eE.mp4", "duration": 150.14999999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy jumps out of the truck and starts to run.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy jumps out of the truck and starts to run.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.8135198354721069, -0.7735597491264343], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.6803196668624878, -0.653679609298706], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002503.pt"}
{"idx": 2505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E_Ci-pAL4eE.mp4", "duration": 150.14999999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the boy hug his grandparents?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy hug his grandparents?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [0.07892107963562012, 0.11888110637664795], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.07892107963562012, 0.11888110637664795], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002505.pt"}
{"idx": 2507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jXSX84eVUCs.mp4", "duration": 148.52, "prompt_text": "Please find the visual event described by the sentence 'When does the red dragon get splashed with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the red dragon get splashed with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [-0.7710745930671692, -0.7441421747207642], "gt_se_raw": [17.0, 18.0], "gt_se_norm": [-0.7710745930671692, -0.7576084136962891], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002507.pt"}
{"idx": 2509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jXSX84eVUCs.mp4", "duration": 148.52, "prompt_text": "Please find the visual event described by the sentence 'When does the girl hide behind a lily pad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl hide behind a lily pad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [-0.39402103424072266, -0.3536224365234375], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.21896040439605713, -0.19202804565429688], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002509.pt"}
{"idx": 2511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jXSX84eVUCs.mp4", "duration": 148.52, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the horse is scared by the three naked men?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the horse is scared by the three naked men?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 128.0], "base_se_norm": [0.589011549949646, 0.7236735820770264], "gt_se_raw": [139.0, 142.0], "gt_se_norm": [0.8718017339706421, 0.9122003316879272], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002511.pt"}
{"idx": 2513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3EBfr6KdnQc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the word 'CAREERS' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the word 'CAREERS' is visible on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.4097805619239807, -0.08188092708587646], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002513.pt"}
{"idx": 2515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3EBfr6KdnQc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does an instructor wearing magnifying loupes appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does an instructor wearing magnifying loupes appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [0.44275856018066406, 0.7706582546234131], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002515.pt"}
{"idx": 2517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3Qp1uk9zSo8.mp4", "duration": 136.59479166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man start opening the envelope and take out the contents?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start opening the envelope and take out the contents?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 27.0], "base_se_norm": [-0.736446738243103, -0.6046701669692993], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.7218049168586731, -0.6485956907272339], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002517.pt"}
{"idx": 2519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3Qp1uk9zSo8.mp4", "duration": 136.59479166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the young man is drinking from a cup with a green straw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the young man is drinking from a cup with a green straw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.1654146909713745, -0.07756364345550537], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.1654146909713745, -0.07756364345550537], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002519.pt"}
{"idx": 2521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3Qp1uk9zSo8.mp4", "duration": 136.59479166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where one man is putting handcuffs on another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where one man is putting handcuffs on another.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 113.0], "base_se_norm": [0.522752046585083, 0.6545287370681763], "gt_se_raw": [104.0, 114.0], "gt_se_norm": [0.522752046585083, 0.669170618057251], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002521.pt"}
{"idx": 2523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DWynJkN5HbQ.mp4", "duration": 139.26412499999998, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man asks the woman where she is from for the second time, emphasizing 'you'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man asks the woman where she is from for the second time, emphasizing 'you'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 46.0], "base_se_norm": [-0.41119080781936646, -0.3393847942352295], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.6696923971176147, -0.64096999168396], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002523.pt"}
{"idx": 2525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DWynJkN5HbQ.mp4", "duration": 139.26412499999998, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman mockingly imitates a British accent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman mockingly imitates a British accent.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 63.0], "base_se_norm": [-0.13832801580429077, -0.0952444076538086], "gt_se_raw": [77.0, 94.0], "gt_se_norm": [0.1058124303817749, 0.3499528169631958], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002525.pt"}
{"idx": 2527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vJG698U2Mvo.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the person in a gorilla costume walking through the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the person in a gorilla costume walking through the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 31.0], "base_se_norm": [-0.41211217641830444, -0.2406449317932129], "gt_se_raw": [23.0, 31.0], "gt_se_norm": [-0.4366075396537781, -0.2406449317932129], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002527.pt"}
{"idx": 2529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vJG698U2Mvo.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the player in the white shirt on the far left walks out of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the player in the white shirt on the far left walks out of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [-0.6080747842788696, -0.5590841770172119], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [-0.46110284328460693, -0.4366075396537781], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002529.pt"}
{"idx": 2531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vJG698U2Mvo.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see two players in white shirts passing the ball to each other on the right side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two players in white shirts passing the ball to each other on the right side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.681560754776001, -0.6080747842788696], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.5590841770172119, -0.5100934505462646], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002531.pt"}
{"idx": 2533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTLCo4PHRAw.mp4", "duration": 36.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the flame from the match transfer to the candle's wick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the flame from the match transfer to the candle's wick?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 24.0], "base_se_norm": [0.11008322238922119, 0.33209991455078125], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.05457913875579834, 0.221091628074646], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002533.pt"}
{"idx": 2535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTLCo4PHRAw.mp4", "duration": 36.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the hand holding the lit match visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the hand holding the lit match visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.3339500427246094, 0.11008322238922119], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.22294169664382935, -0.0009250640869140625], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002535.pt"}
{"idx": 2537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aMtY7TqY_eQ.mp4", "duration": 12.5, "prompt_text": "Please find the visual event described by the sentence 'When is the food taken out of the microwave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the food taken out of the microwave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.8400000333786011, -0.5199999809265137], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.8400000333786011, -0.5199999809265137], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002537.pt"}
{"idx": 2539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aMtY7TqY_eQ.mp4", "duration": 12.5, "prompt_text": "Please find the visual event described by the sentence 'At what point is the microwave door closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the microwave door closed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.6800000071525574, -0.36000001430511475], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.5199999809265137, -0.36000001430511475], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002539.pt"}
{"idx": 2541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aMtY7TqY_eQ.mp4", "duration": 12.5, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the filling of the pastry is revealed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the filling of the pastry is revealed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.19999998807907104, 0.2799999713897705], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.19999998807907104, 0.2799999713897705], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002541.pt"}
{"idx": 2543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/at-2rA44S-M.mp4", "duration": 52.452400000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man is working on the airplane's engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man is working on the airplane's engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.1611441969871521, -0.00862497091293335], "gt_se_raw": [22.0, 26.0], "gt_se_norm": [-0.1611441969871521, -0.00862497091293335], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002543.pt"}
{"idx": 2545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/at-2rA44S-M.mp4", "duration": 52.452400000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when three men are standing together and talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when three men are standing together and talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 36.0], "base_se_norm": [0.029504895210266113, 0.3726731538772583], "gt_se_raw": [27.0, 41.0], "gt_se_norm": [0.029504895210266113, 0.5633221864700317], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002545.pt"}
{"idx": 2547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/woWiyBgp5cs.mp4", "duration": 66.29956666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the pink egg toy placed in the flowerpot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the pink egg toy placed in the flowerpot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 58.0], "base_se_norm": [0.29714274406433105, 0.7496342658996582], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.8491694331169128, -0.8190033435821533], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002547.pt"}
{"idx": 2549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/woWiyBgp5cs.mp4", "duration": 66.29956666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy successfully finds and picks up the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy successfully finds and picks up the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 59.0], "base_se_norm": [0.29714274406433105, 0.7798004150390625], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [0.6591359376907349, 0.7496342658996582], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002549.pt"}
{"idx": 2551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/woWiyBgp5cs.mp4", "duration": 66.29956666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man hug the boy to praise him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man hug the boy to praise him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 39.0], "base_se_norm": [-0.0950167179107666, 0.17647826671600342], "gt_se_raw": [64.0, 65.0], "gt_se_norm": [0.9306309223175049, 0.9607970714569092], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002551.pt"}
{"idx": 2553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvhlK0WGBr4.mp4", "duration": 179.5, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the cracked, dry earth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the cracked, dry earth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 21.0], "base_se_norm": [-0.8662952780723572, -0.7660167217254639], "gt_se_raw": [12.0, 20.0], "gt_se_norm": [-0.8662952780723572, -0.7771587371826172], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002553.pt"}
{"idx": 2555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvhlK0WGBr4.mp4", "duration": 179.5, "prompt_text": "Please find the visual event described by the sentence 'Find the time range when a large, striped planet appears in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range when a large, striped planet appears in the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 129.0], "base_se_norm": [0.3370473384857178, 0.43732595443725586], "gt_se_raw": [120.0, 128.0], "gt_se_norm": [0.3370473384857178, 0.426183819770813], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002555.pt"}
{"idx": 2557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Qdwl-L_PR8.mp4", "duration": 160.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see people riding horses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see people riding horses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 68.0], "base_se_norm": [-0.25434964895248413, -0.15492957830429077], "gt_se_raw": [60.0, 67.0], "gt_se_norm": [-0.25434964895248413, -0.16735708713531494], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002557.pt"}
{"idx": 2559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Qdwl-L_PR8.mp4", "duration": 160.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do a group of people pose for a photo by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do a group of people pose for a photo by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.41590720415115356, -0.37862467765808105], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.41590720415115356, -0.37862467765808105], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002559.pt"}
{"idx": 2561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Qdwl-L_PR8.mp4", "duration": 160.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame for the person stacking and climbing crates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame for the person stacking and climbing crates?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 103.0], "base_se_norm": [0.09362053871154785, 0.2800331115722656], "gt_se_raw": [89.0, 103.0], "gt_se_norm": [0.10604798793792725, 0.2800331115722656], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002561.pt"}
{"idx": 2563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hsm4poTWjMs.mp4", "duration": 252.25199999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman in the purple outfit points directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman in the purple outfit points directly at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [-0.41328513622283936, -0.38157081604003906], "gt_se_raw": [74.0, 75.0], "gt_se_norm": [-0.41328513622283936, -0.4053565263748169], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002563.pt"}
{"idx": 2565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hsm4poTWjMs.mp4", "duration": 252.25199999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman in leopard print does a split between two chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman in leopard print does a split between two chairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 126.0], "base_se_norm": [-0.024784743785858154, -0.0009989738464355469], "gt_se_raw": [124.0, 126.0], "gt_se_norm": [-0.016856133937835693, -0.0009989738464355469], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002565.pt"}
{"idx": 2567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQP1duR2tf4.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman in white gazes up at the ceiling paintings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman in white gazes up at the ceiling paintings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 31.0], "base_se_norm": [-0.8876625299453735, -0.5025055408477783], "gt_se_raw": [11.0, 30.0], "gt_se_norm": [-0.8234696984291077, -0.5185537338256836], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002567.pt"}
{"idx": 2569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQP1duR2tf4.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman angrily pushes and strikes the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman angrily pushes and strikes the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.1815413236618042, -0.10130029916763306], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.1333966851234436, -0.10130029916763306], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002569.pt"}
{"idx": 2571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rQP1duR2tf4.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Show the final scene where the woman turns and walks out of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the final scene where the woman turns and walks out of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 94.0], "base_se_norm": [0.42829060554504395, 0.5085316896438599], "gt_se_raw": [88.0, 93.0], "gt_se_norm": [0.41224241256713867, 0.4924834966659546], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002571.pt"}
{"idx": 2573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oVS7kHYlLBc.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment someone fills the filter basket with ground coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment someone fills the filter basket with ground coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.7751124501228333, -0.6501749157905579], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.7751124501228333, -0.6501749157905579], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002573.pt"}
{"idx": 2575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oVS7kHYlLBc.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'At what time does the coffee brew and rise into the upper pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the coffee brew and rise into the upper pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 46.0], "base_se_norm": [-0.12543731927871704, 0.14942526817321777], "gt_se_raw": [35.0, 46.0], "gt_se_norm": [-0.12543731927871704, 0.14942526817321777], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002575.pt"}
{"idx": 2577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R-t2Zbzl5mk.mp4", "duration": 56.72332222222222, "prompt_text": "Please find the visual event described by the sentence 'When are the artichoke hearts added to the spinach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the artichoke hearts added to the spinach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.5768936276435852, -0.4711170196533203], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.5768936276435852, -0.5063759088516235], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002577.pt"}
{"idx": 2579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R-t2Zbzl5mk.mp4", "duration": 56.72332222222222, "prompt_text": "Please find the visual event described by the sentence 'At what point are cheddar and mozzarella added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are cheddar and mozzarella added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 35.0], "base_se_norm": [0.0930248498916626, 0.23406028747558594], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [0.0930248498916626, 0.23406028747558594], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002579.pt"}
{"idx": 2581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R-t2Zbzl5mk.mp4", "duration": 56.72332222222222, "prompt_text": "Please find the visual event described by the sentence 'When can we see the final, baked dish with a golden-brown crust?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the final, baked dish with a golden-brown crust?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 54.0], "base_se_norm": [0.657166600227356, 0.903978705406189], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [0.6924254894256592, 0.7982021570205688], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002581.pt"}
{"idx": 2583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oP0OgYPSXSE.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker state that building an emotional relationship is 'foundational'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker state that building an emotional relationship is 'foundational'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [0.3020516633987427, 0.41626667976379395], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.6116688251495361, -0.5431398153305054], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002583.pt"}
{"idx": 2585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oP0OgYPSXSE.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman explains how her son wouldn't bother communicating if others didn't make an effort.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman explains how her son wouldn't bother communicating if others didn't make an effort.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 63.0], "base_se_norm": [0.09646451473236084, 0.4391096830368042], "gt_se_raw": [50.0, 58.0], "gt_se_norm": [0.14215052127838135, 0.32489466667175293], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002585.pt"}
{"idx": 2587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkEEO_TLNbM.mp4", "duration": 65.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where people are watching two scuba divers inside a large aquarium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where people are watching two scuba divers inside a large aquarium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 40.0], "base_se_norm": [0.1538461446762085, 0.21457481384277344], "gt_se_raw": [57.0, 58.0], "gt_se_norm": [0.730769157409668, 0.7611335515975952], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002587.pt"}
{"idx": 2589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkEEO_TLNbM.mp4", "duration": 65.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does a young man in a suit and tie give a speech at a podium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a young man in a suit and tie give a speech at a podium?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 53.0], "base_se_norm": [0.5182185173034668, 0.6093116998672485], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.4878542423248291, 0.5789473056793213], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002589.pt"}
{"idx": 2591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkEEO_TLNbM.mp4", "duration": 65.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time range showing a turkey vulture standing next to a skeleton in a desert exhibit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range showing a turkey vulture standing next to a skeleton in a desert exhibit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [-0.36234819889068604, -0.3016194701194763], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [-0.3319838047027588, -0.3016194701194763], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002591.pt"}
{"idx": 2593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mize843q3iM.mp4", "duration": 30.113416666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a student has a white coat placed on her shoulders during a ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a student has a white coat placed on her shoulders during a ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 8.0], "base_se_norm": [-0.5350909233093262, -0.4686753749847412], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.6679221391677856, -0.5350909233093262], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002593.pt"}
{"idx": 2595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mize843q3iM.mp4", "duration": 30.113416666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where two students are practicing on a mannequin in a hospital bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where two students are practicing on a mannequin in a hospital bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.5275583267211914, 0.6603894233703613], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.5275583267211914, 0.6603894233703613], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002595.pt"}
{"idx": 2597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hzkbhZhgHAU.mp4", "duration": 132.5324, "prompt_text": "Please find the visual event described by the sentence 'When can we observe two octopuses dancing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe two octopuses dancing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 12.0], "base_se_norm": [-0.9396374225616455, -0.8189122080802917], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.9245467185974121, -0.8641841411590576], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002597.pt"}
{"idx": 2599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hzkbhZhgHAU.mp4", "duration": 132.5324, "prompt_text": "Please find the visual event described by the sentence 'When does the orange octopus grab the driver's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the orange octopus grab the driver's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [-0.15492361783981323, -0.10965162515640259], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [-0.09456098079681396, -0.06437969207763672], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002599.pt"}
{"idx": 2601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hzkbhZhgHAU.mp4", "duration": 132.5324, "prompt_text": "Please find the visual event described by the sentence 'When does the truck crash into the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the truck crash into the sea?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 114.0], "base_se_norm": [0.6901527643203735, 0.7203341722488403], "gt_se_raw": [112.0, 114.0], "gt_se_norm": [0.6901527643203735, 0.7203341722488403], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002601.pt"}
{"idx": 2603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JvQcabZ1zrk.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'When does the new dog appear on the kitchen floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the new dog appear on the kitchen floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 120.0], "base_se_norm": [0.4956315755844116, 0.574349045753479], "gt_se_raw": [113.0, 121.0], "gt_se_norm": [0.4825119972229004, 0.5874686241149902], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002603.pt"}
{"idx": 2605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JvQcabZ1zrk.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the 'Privacy Window' feature is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the 'Privacy Window' feature is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 85.0], "base_se_norm": [-0.01603180170059204, 0.11516392230987549], "gt_se_raw": [75.0, 83.0], "gt_se_norm": [-0.01603180170059204, 0.08892476558685303], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002605.pt"}
{"idx": 2607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p4cc4LkJFjg.mp4", "duration": 201.23436666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the text '#RaceAnd GENDER' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text '#RaceAnd GENDER' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 8.0], "base_se_norm": [-0.9304293990135193, -0.9204907417297363], "gt_se_raw": [7.0, 8.0], "gt_se_norm": [-0.9304293990135193, -0.9204907417297363], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002607.pt"}
{"idx": 2609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p4cc4LkJFjg.mp4", "duration": 201.23436666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time does the video ask the question 'Have there been times when your identities have been ignored?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the video ask the question 'Have there been times when your identities have been ignored?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [-0.622330904006958, -0.5825762748718262], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.612392246723175, -0.5825762748718262], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002609.pt"}
{"idx": 2611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p4cc4LkJFjg.mp4", "duration": 201.23436666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when a map of the US, Mexico, and Canada is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when a map of the US, Mexico, and Canada is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 122.0], "base_se_norm": [0.09325265884399414, 0.21251654624938965], "gt_se_raw": [110.0, 122.0], "gt_se_norm": [0.09325265884399414, 0.21251654624938965], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002611.pt"}
{"idx": 2613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pEQ7QttNnxE.mp4", "duration": 25.4, "prompt_text": "Please find the visual event described by the sentence 'Show the interval where the person is inspecting a blue plastic tool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the interval where the person is inspecting a blue plastic tool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9212598204612732, -0.6062992215156555], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.8425197005271912, -0.6850394010543823], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002613.pt"}
{"idx": 2615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pEQ7QttNnxE.mp4", "duration": 25.4, "prompt_text": "Please find the visual event described by the sentence 'When does the person slide the red shape across the illuminated circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person slide the red shape across the illuminated circle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 23.0], "base_se_norm": [-0.4488188624382019, 0.8110235929489136], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.21259844303131104, 0.10236227512359619], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002615.pt"}
{"idx": 2617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xZfZ-HB2yJI.mp4", "duration": 45.845800000000004, "prompt_text": "Please find the visual event described by the sentence 'When does the woman give a thumbs up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman give a thumbs up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.8691264986991882, -0.7382529973983765], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.8255020380020142, -0.7382529973983765], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002617.pt"}
{"idx": 2619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xZfZ-HB2yJI.mp4", "duration": 45.845800000000004, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two people holding hands at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two people holding hands at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.7382529973983765], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8255020380020142], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002619.pt"}
{"idx": 2621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xZfZ-HB2yJI.mp4", "duration": 45.845800000000004, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the two individuals are holding hands for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the two individuals are holding hands for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 45.0], "base_se_norm": [0.875853419303894, 0.9631024599075317], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [0.875853419303894, 0.9631024599075317], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002621.pt"}
{"idx": 2623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2kUMAA9yZgk.mp4", "duration": 131.88, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl hide behind a log and then a little monkey appears?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl hide behind a log and then a little monkey appears?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 62.0], "base_se_norm": [-0.39338797330856323, -0.05975133180618286], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.43888384103775024, -0.36305737495422363], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002623.pt"}
{"idx": 2625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2kUMAA9yZgk.mp4", "duration": 131.88, "prompt_text": "Please find the visual event described by the sentence 'When does the little monkey's eyes suddenly glow blue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little monkey's eyes suddenly glow blue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [-0.014255404472351074, 0.031240463256835938], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [-0.014255404472351074, 0.031240463256835938], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002625.pt"}
{"idx": 2627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/36OitAr_YQY.mp4", "duration": 47.98126666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the plaid shirt start talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the plaid shirt start talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, -0.4164389371871948], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8332682847976685], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002627.pt"}
{"idx": 2629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/36OitAr_YQY.mp4", "duration": 47.98126666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the woman in the white dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the woman in the white dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.20802420377731323, 0.0003904104232788086], "gt_se_raw": [20.0, 25.0], "gt_se_norm": [-0.16634130477905273, 0.04207336902618408], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002629.pt"}
{"idx": 2631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4wEA99kE8c.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the segment that indicates the location of Niagara Falls on the border.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the segment that indicates the location of Niagara Falls on the border.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.7102615833282471, -0.6136820912361145], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.7102615833282471, -0.6136820912361145], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002631.pt"}
{"idx": 2633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4wEA99kE8c.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video explain that the falls were created by glacier activity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video explain that the falls were created by glacier activity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 41.0], "base_se_norm": [-0.227364182472229, -0.010060369968414307], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [-0.227364182472229, -0.010060369968414307], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002633.pt"}
{"idx": 2635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4wEA99kE8c.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a daytime rainbow, or solar bow, appears over the falls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a daytime rainbow, or solar bow, appears over the falls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 71.0], "base_se_norm": [0.545271635055542, 0.7142856121063232], "gt_se_raw": [64.0, 71.0], "gt_se_norm": [0.545271635055542, 0.7142856121063232], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002635.pt"}
{"idx": 2637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4nJYJ-QcABs.mp4", "duration": 203.36983333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a girl in a purple dress is seen up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a girl in a purple dress is seen up close.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 49.0], "base_se_norm": [-0.7344738841056824, -0.5181192755699158], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.7443081736564636, -0.6951366662979126], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002637.pt"}
{"idx": 2639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4nJYJ-QcABs.mp4", "duration": 203.36983333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a boy think 'Observe without judgement' inside a thought bubble?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a boy think 'Observe without judgement' inside a thought bubble?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 63.0], "base_se_norm": [-0.439444899559021, -0.38043904304504395], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.439444899559021, -0.38043904304504395], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002639.pt"}
{"idx": 2641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FatA6Pyw_D8.mp4", "duration": 72.0, "prompt_text": "Please find the visual event described by the sentence 'When does the dog first appear and offer a ball to the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog first appear and offer a ball to the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.7777777910232544, -0.6388888955116272], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.7777777910232544, -0.6666666269302368], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002641.pt"}
{"idx": 2643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FatA6Pyw_D8.mp4", "duration": 72.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat throw the frisbee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat throw the frisbee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [-0.05555558204650879, 0.0], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [0.05555558204650879, 0.11111116409301758], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002643.pt"}
{"idx": 2645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FatA6Pyw_D8.mp4", "duration": 72.0, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the dog crashes through the fence and returns with the stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the dog crashes through the fence and returns with the stick.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 61.0], "base_se_norm": [0.5555555820465088, 0.6944444179534912], "gt_se_raw": [52.0, 59.0], "gt_se_norm": [0.4444444179534912, 0.6388888359069824], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002645.pt"}
{"idx": 2647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZHlfQSL7dJ8.mp4", "duration": 57.22383333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a man with tattoos on his arm is speaking to a group of people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a man with tattoos on his arm is speaking to a group of people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.40584200620651245, -0.2660400867462158], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.40584200620651245, -0.2660400867462158], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002647.pt"}
{"idx": 2649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZHlfQSL7dJ8.mp4", "duration": 57.22383333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a group of people are sitting in a circle and engaging in a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a group of people are sitting in a circle and engaging in a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 39.0], "base_se_norm": [0.15336549282073975, 0.3630683422088623], "gt_se_raw": [33.0, 39.0], "gt_se_norm": [0.15336549282073975, 0.3630683422088623], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002649.pt"}
{"idx": 2651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-TQpVg4vcrk.mp4", "duration": 60.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the camouflage jacket enter the room to meet the other man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the camouflage jacket enter the room to meet the other man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.8002219796180725, -0.6670366525650024], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.8002219796180725, -0.7336292862892151], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002651.pt"}
{"idx": 2653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-TQpVg4vcrk.mp4", "duration": 60.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the cap put his hand on the other man's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the cap put his hand on the other man's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.23418420553207397, -0.16759157180786133], "gt_se_raw": [44.0, 45.0], "gt_se_norm": [0.46503889560699463, 0.49833524227142334], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002653.pt"}
{"idx": 2655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-TQpVg4vcrk.mp4", "duration": 60.06666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the two men hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the two men hug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 50.0], "base_se_norm": [0.5982241630554199, 0.6648168563842773], "gt_se_raw": [49.0, 50.0], "gt_se_norm": [0.6315205097198486, 0.6648168563842773], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002655.pt"}
{"idx": 2657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f25JmqQMYow.mp4", "duration": 115.6155, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the blue shirt speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the blue shirt speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 25.0], "base_se_norm": [-0.7232204675674438, -0.5675320625305176], "gt_se_raw": [16.0, 25.0], "gt_se_norm": [-0.7232204675674438, -0.5675320625305176], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002657.pt"}
{"idx": 2659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f25JmqQMYow.mp4", "duration": 115.6155, "prompt_text": "Please find the visual event described by the sentence 'At what time does the gorilla walk past a small waterfall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the gorilla walk past a small waterfall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [-0.2907525300979614, -0.22155767679214478], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.2907525300979614, -0.2561551332473755], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002659.pt"}
{"idx": 2661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5hpBAn5lQPs.mp4", "duration": 57.057, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the opening title 'Physical Support & Need for Rest' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the opening title 'Physical Support & Need for Rest' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.789683997631073], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.789683997631073], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002661.pt"}
{"idx": 2663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5hpBAn5lQPs.mp4", "duration": 57.057, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker explain that the boy's leaning impacts his ability to see?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker explain that the boy's leaning impacts his ability to see?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [-0.43915730714797974, -0.22884130477905273], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.40410465002059937, -0.22884130477905273], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002663.pt"}
{"idx": 2665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5hpBAn5lQPs.mp4", "duration": 57.057, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention giving the boy 'time to relax out of his wheelchair'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention giving the boy 'time to relax out of his wheelchair'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 56.0], "base_se_norm": [0.5072647333145142, 0.9629493951797485], "gt_se_raw": [46.0, 50.0], "gt_se_norm": [0.6124227046966553, 0.7526333332061768], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002665.pt"}
{"idx": 2667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fTcSVQJ2h8g.mp4", "duration": 184.35083333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a police officer interacting with only one of the drivers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a police officer interacting with only one of the drivers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.2514272928237915, -0.2080317735671997], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [-0.2514272928237915, -0.21888065338134766], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002667.pt"}
{"idx": 2669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fTcSVQJ2h8g.mp4", "duration": 184.35083333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time are the resumes shown to be identical except for the names 'Dante' and 'David'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the resumes shown to be identical except for the names 'Dante' and 'David'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.7613246440887451, -0.6962313652038574], "gt_se_raw": [19.0, 27.0], "gt_se_norm": [-0.793871283531189, -0.7070802450180054], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002669.pt"}
{"idx": 2671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_QG77WomHo.mp4", "duration": 159.159, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the artist is drawing with a black marker in his studio.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the artist is drawing with a black marker in his studio.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 18.0], "base_se_norm": [-0.8994715809822083, -0.7738111019134521], "gt_se_raw": [9.0, 17.0], "gt_se_norm": [-0.8869055509567261, -0.7863771319389343], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002671.pt"}
{"idx": 2673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_QG77WomHo.mp4", "duration": 159.159, "prompt_text": "Please find the visual event described by the sentence 'Show the scene with various graffiti art on walls and a truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene with various graffiti art on walls and a truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 64.0], "base_se_norm": [-0.32143324613571167, -0.19577276706695557], "gt_se_raw": [54.0, 63.0], "gt_se_norm": [-0.32143324613571167, -0.20833879709243774], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002673.pt"}
{"idx": 2675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_QG77WomHo.mp4", "duration": 159.159, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the artist begins spray-painting the wall mural.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the artist begins spray-painting the wall mural.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 112.0], "base_se_norm": [0.24403905868530273, 0.40739762783050537], "gt_se_raw": [106.0, 112.0], "gt_se_norm": [0.33200132846832275, 0.40739762783050537], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002675.pt"}
{"idx": 2677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ywzm6kvDo7o.mp4", "duration": 135.76896666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when fish with yellow tails are swimming.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when fish with yellow tails are swimming.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 40.0], "base_se_norm": [-0.4696873426437378, -0.4107637405395508], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [-0.48441827297210693, -0.4107637405395508], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002677.pt"}
{"idx": 2679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ywzm6kvDo7o.mp4", "duration": 135.76896666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up of the giant sea bass's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up of the giant sea bass's face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 88.0], "base_se_norm": [0.19320344924926758, 0.296319842338562], "gt_se_raw": [81.0, 88.0], "gt_se_norm": [0.19320344924926758, 0.296319842338562], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002679.pt"}
{"idx": 2681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SHR0GuIf5eM.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe two men working on an engine together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe two men working on an engine together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002681.pt"}
{"idx": 2683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SHR0GuIf5eM.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a man is inspecting the inside of a large jet engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a man is inspecting the inside of a large jet engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002683.pt"}
{"idx": 2685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SHR0GuIf5eM.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'CAREERS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'CAREERS' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.4097805619239807, -0.08188092708587646], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002685.pt"}
{"idx": 2687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XlUMyKgmy6k.mp4", "duration": 133.24, "prompt_text": "Please find the visual event described by the sentence 'When does the sun appear next to the Earth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the sun appear next to the Earth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.6247373223304749, -0.5797058343887329], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.6247373223304749, -0.5947163105010986], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002687.pt"}
{"idx": 2689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XlUMyKgmy6k.mp4", "duration": 133.24, "prompt_text": "Please find the visual event described by the sentence 'When is there an intervention to stop someone from breaking necks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is there an intervention to stop someone from breaking necks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.5196638107299805, -0.459621787071228], "gt_se_raw": [30.0, 37.0], "gt_se_norm": [-0.5496847629547119, -0.44461125135421753], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002689.pt"}
{"idx": 2691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/19143I5f3Lg.mp4", "duration": 150.88406666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the poster for the documentary 'Who Cares About Kelsey?' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the poster for the documentary 'Who Cares About Kelsey?' appear on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9602343440055847, -0.9072135090827942], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9602343440055847, -0.9072135090827942], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002691.pt"}
{"idx": 2693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/19143I5f3Lg.mp4", "duration": 150.88406666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video state that 60% of students with emotional and behavioral disabilities are likely to drop out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video state that 60% of students with emotional and behavioral disabilities are likely to drop out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 71.0], "base_se_norm": [-0.17817693948745728, -0.05888009071350098], "gt_se_raw": [63.0, 71.0], "gt_se_norm": [-0.16492176055908203, -0.05888009071350098], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002693.pt"}
{"idx": 2695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/19143I5f3Lg.mp4", "duration": 150.88406666666668, "prompt_text": "Please find the visual event described by the sentence 'When are the statistics about the reduced dropout and discipline problems shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the statistics about the reduced dropout and discipline problems shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 140.0], "base_se_norm": [0.7364325523376465, 0.8557294607162476], "gt_se_raw": [131.0, 141.0], "gt_se_norm": [0.7364325523376465, 0.8689846992492676], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002695.pt"}
{"idx": 2697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Hf_2ilitep0.mp4", "duration": 117.8177, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the teacher demonstrates how to form a chord using the hand-over-hand technique.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the teacher demonstrates how to form a chord using the hand-over-hand technique.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 92.0], "base_se_norm": [0.239202618598938, 0.5617347955703735], "gt_se_raw": [46.0, 58.0], "gt_se_norm": [-0.21913260221481323, -0.015428125858306885], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002697.pt"}
{"idx": 2699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Hf_2ilitep0.mp4", "duration": 117.8177, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the instructor uses his hand to demonstrate an up-and-down strumming motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the instructor uses his hand to demonstrate an up-and-down strumming motion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 42.0], "base_se_norm": [-0.4058617949485779, -0.28703415393829346], "gt_se_raw": [106.0, 109.0], "gt_se_norm": [0.7993900775909424, 0.8503161668777466], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002699.pt"}
{"idx": 2701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1O-HcBYpNM0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the woman use a wooden block to rub the girl's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman use a wooden block to rub the girl's hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [0.4525054693222046, 0.9366739988327026], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [0.4525054693222046, 0.9366739988327026], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002701.pt"}
{"idx": 2703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1O-HcBYpNM0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman help the girl rub her hands together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman help the girl rub her hands together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.06517064571380615], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.3221641182899475, 0.06517064571380615], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002703.pt"}
{"idx": 2705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1O-HcBYpNM0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [0.7430065870285034, 0.9366739988327026], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.7430065870285034, 0.9366739988327026], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002705.pt"}
{"idx": 2707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kQQaEo8MqDQ.mp4", "duration": 261.96666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a woman in a purple dress touches the artwork.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a woman in a purple dress touches the artwork.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.732790470123291, -0.7098867893218994], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [-0.7251558899879456, -0.7098867893218994], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002707.pt"}
{"idx": 2709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kQQaEo8MqDQ.mp4", "duration": 261.96666666666664, "prompt_text": "Please find the visual event described by the sentence 'When are two men closely examining a painting together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men closely examining a painting together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 143.0], "base_se_norm": [-0.2441786527633667, 0.09174191951751709], "gt_se_raw": [99.0, 149.0], "gt_se_norm": [-0.2441786527633667, 0.1375492811203003], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002709.pt"}
{"idx": 2711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CKwUNBEFI0E.mp4", "duration": 156.768, "prompt_text": "Please find the visual event described by the sentence 'Show the front view of the jet while it is stationary on the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the front view of the jet while it is stationary on the runway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 34.0], "base_se_norm": [-0.6427842378616333, -0.5662380456924438], "gt_se_raw": [29.0, 34.0], "gt_se_norm": [-0.6300265789031982, -0.5662380456924438], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002711.pt"}
{"idx": 2713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CKwUNBEFI0E.mp4", "duration": 156.768, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the model jet accelerates down the runway and lifts into the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the model jet accelerates down the runway and lifts into the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 77.0], "base_se_norm": [-0.08144521713256836, -0.017656683921813965], "gt_se_raw": [70.0, 76.0], "gt_se_norm": [-0.10696065425872803, -0.0304144024848938], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002713.pt"}
{"idx": 2715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CKwUNBEFI0E.mp4", "duration": 156.768, "prompt_text": "Please find the visual event described by the sentence 'When does the jet touch down on the runway after flying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the jet touch down on the runway after flying?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 155.0], "base_se_norm": [0.9008980989456177, 0.9774442911148071], "gt_se_raw": [149.0, 152.0], "gt_se_norm": [0.9008980989456177, 0.9391711950302124], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002715.pt"}
{"idx": 2717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4XwDbm_37cw.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'Locate the interview with the elderly couple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the interview with the elderly couple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 126.0], "base_se_norm": [0.3401981592178345, 0.5492199659347534], "gt_se_raw": [109.0, 126.0], "gt_se_norm": [0.3401981592178345, 0.5492199659347534], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002717.pt"}
{"idx": 2719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4XwDbm_37cw.mp4", "duration": 162.6625, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where an instructor is teaching a class of three students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 126.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where an instructor is teaching a class of three students.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 126.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [126.0, 130.0], "base_se_norm": [0.5492199659347534, 0.5984015464782715], "gt_se_raw": [126.0, 130.0], "gt_se_norm": [0.5492199659347534, 0.5984015464782715], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002719.pt"}
{"idx": 2721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DMCSP73Rq4I.mp4", "duration": 57.182125, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of the 'BARGAINS OF THE MONTH' flyer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of the 'BARGAINS OF THE MONTH' flyer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.6852163076400757, -0.44038456678390503], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.6852163076400757, -0.4753605127334595], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002721.pt"}
{"idx": 2723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DMCSP73Rq4I.mp4", "duration": 57.182125, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man is crumpling the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man is crumpling the paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [-0.26550471782684326, -0.19555282592773438], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.23052877187728882, -0.16057682037353516], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002723.pt"}
{"idx": 2725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DMCSP73Rq4I.mp4", "duration": 57.182125, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the end screen with the text 'SUBSCRIBE TO DAD' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the end screen with the text 'SUBSCRIBE TO DAD' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 56.0], "base_se_norm": [0.32908666133880615, 0.9586540460586548], "gt_se_raw": [37.0, 57.0], "gt_se_norm": [0.29411065578460693, 0.993630051612854], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002725.pt"}
{"idx": 2727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGJ_td3ZUIU.mp4", "duration": 196.62976666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the speaker lists her three identities as Black, Woman, and Queer with corresponding text overlays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the speaker lists her three identities as Black, Woman, and Queer with corresponding text overlays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.8474289774894714, -0.8067433834075928], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.8474289774894714, -0.8067433834075928], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002727.pt"}
{"idx": 2729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGJ_td3ZUIU.mp4", "duration": 196.62976666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the video display the definition of 'Heteronormativity'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display the definition of 'Heteronormativity'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 127.0], "base_se_norm": [0.23073947429656982, 0.29176783561706543], "gt_se_raw": [122.0, 127.0], "gt_se_norm": [0.24091088771820068, 0.29176783561706543], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002729.pt"}
{"idx": 2731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hIC7XFLhJ7A.mp4", "duration": 106.37293333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a person in a black hooded costume is manipulating a puppet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a person in a black hooded costume is manipulating a puppet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 23.0], "base_se_norm": [-0.7743786573410034, -0.5675591230392456], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.6803698539733887, -0.5675591230392456], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002731.pt"}
{"idx": 2733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hIC7XFLhJ7A.mp4", "duration": 106.37293333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman in a red coat is searching for something in her bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman in a red coat is searching for something in her bag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 55.0], "base_se_norm": [-0.041109442710876465, 0.03409767150878906], "gt_se_raw": [49.0, 53.0], "gt_se_norm": [-0.07871299982070923, -0.003505885601043701], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002733.pt"}
{"idx": 2735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hIC7XFLhJ7A.mp4", "duration": 106.37293333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time when a man hugs a woman from behind while she looks at a piece of cloth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when a man hugs a woman from behind while she looks at a piece of cloth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [-0.26673072576522827, -0.22912722826004028], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.28553253412246704, -0.22912722826004028], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002735.pt"}
{"idx": 2737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/545Tp9xXEIQ.mp4", "duration": 46.08770833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the yellow minion offers a banana to the purple minion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the yellow minion offers a banana to the purple minion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.34906721115112305, -0.04529857635498047], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.21888065338134766, -0.08869409561157227], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002737.pt"}
{"idx": 2739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/545Tp9xXEIQ.mp4", "duration": 46.08770833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the purple minion chases the yellow minion across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the purple minion chases the yellow minion across the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 45.0], "base_se_norm": [0.5188431739807129, 0.9527983665466309], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [0.5188431739807129, 0.7358207702636719], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002739.pt"}
{"idx": 2741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fiKThScYotw.mp4", "duration": 60.3, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy walking down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy walking down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.3698175549507141, -0.27031511068344116], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.3698175549507141, -0.27031511068344116], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002741.pt"}
{"idx": 2743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fiKThScYotw.mp4", "duration": 60.3, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the boy is walking on the sidewalk crossing a white painted rectangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the boy is walking on the sidewalk crossing a white painted rectangle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [-0.038142621517181396, 0.06135988235473633], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.038142621517181396, 0.06135988235473633], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002743.pt"}
{"idx": 2745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fiKThScYotw.mp4", "duration": 60.3, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the boy's face as he smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the boy's face as he smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 54.0], "base_se_norm": [0.7247097492218018, 0.7910448312759399], "gt_se_raw": [52.0, 53.0], "gt_se_norm": [0.7247097492218018, 0.7578773498535156], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002745.pt"}
{"idx": 2747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_yQABXDAy2Q.mp4", "duration": 32.098733333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the girl describes how Braille is read by feeling the dots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the girl describes how Braille is read by feeling the dots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.127691388130188, 0.18384742736816406], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [-0.003075897693634033, 0.24615514278411865], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002747.pt"}
{"idx": 2749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_yQABXDAy2Q.mp4", "duration": 32.098733333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the girl is asked if Braille is easy to learn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the girl is asked if Braille is easy to learn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [0.8069249391555786, 0.9315404891967773], "gt_se_raw": [29.0, 30.0], "gt_se_norm": [0.8069249391555786, 0.8692327737808228], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002749.pt"}
{"idx": 2751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NWIVHL9bPPU.mp4", "duration": 146.27944611277945, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a group of students is studying together around a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a group of students is studying together around a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.9453101754188538, -0.9042927622795105], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9589825868606567, -0.9042927622795105], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002751.pt"}
{"idx": 2753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NWIVHL9bPPU.mp4", "duration": 146.27944611277945, "prompt_text": "Please find the visual event described by the sentence 'Identify the part of the video that shows a black and white image of a man in a suit, holding his head with question marks floating around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the part of the video that shows a black and white image of a man in a suit, holding his head with question marks floating around.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 71.0], "base_se_norm": [-0.12496250867843628, -0.029255330562591553], "gt_se_raw": [64.0, 71.0], "gt_se_norm": [-0.12496250867843628, -0.029255330562591553], "start_token_span": [84, 88], "end_token_span": [90, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002753.pt"}
{"idx": 2755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NWIVHL9bPPU.mp4", "duration": 146.27944611277945, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the video show the phrase 'MAKE THINGS BETTER' being written on a chalkboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the video show the phrase 'MAKE THINGS BETTER' being written on a chalkboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 141.0], "base_se_norm": [0.8731271028518677, 0.9278169870376587], "gt_se_raw": [137.0, 142.0], "gt_se_norm": [0.8731271028518677, 0.9414893388748169], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002755.pt"}
{"idx": 2757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p1a5oijzFaY.mp4", "duration": 111.54476666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person in the blue shirt smile and bring their hand up to their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person in the blue shirt smile and bring their hand up to their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.5517494678497314, -0.4441693425178528], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.5517494678497314, -0.4800294041633606], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002757.pt"}
{"idx": 2759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p1a5oijzFaY.mp4", "duration": 111.54476666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the person in the blue shirt touch the straw on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in the blue shirt touch the straw on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 59.0], "base_se_norm": [-0.04970890283584595, 0.057871222496032715], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [-0.04970890283584595, -0.01384890079498291], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002759.pt"}
{"idx": 2761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODDZPV-Avfw.mp4", "duration": 345.64, "prompt_text": "Please find the visual event described by the sentence 'When are 'lone sharks' and 'burglars' depicted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are 'lone sharks' and 'burglars' depicted?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.9479227066040039, -0.9363499879837036], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.9479227066040039, -0.9363499879837036], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002761.pt"}
{"idx": 2763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODDZPV-Avfw.mp4", "duration": 345.64, "prompt_text": "Please find the visual event described by the sentence 'At what moment does a man in a white shirt and tie point at a board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does a man in a white shirt and tie point at a board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [-0.5660224556922913, -0.5370906591415405], "gt_se_raw": [76.0, 79.0], "gt_se_norm": [-0.5602360963821411, -0.5428769588470459], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002763.pt"}
{"idx": 2765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODDZPV-Avfw.mp4", "duration": 345.64, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a sad, blue-colored character is shown against a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 184.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a sad, blue-colored character is shown against a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 184.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [184.0, 199.0], "base_se_norm": [0.06469154357910156, 0.15148699283599854], "gt_se_raw": [200.0, 206.0], "gt_se_norm": [0.15727341175079346, 0.19199156761169434], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002765.pt"}
{"idx": 2767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vI_B7dtF7Q0.mp4", "duration": 297.04675, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the orange is shown wearing a tuxedo and looking through binoculars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the orange is shown wearing a tuxedo and looking through binoculars.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 105.0], "base_se_norm": [-0.3738359212875366, -0.29304057359695435], "gt_se_raw": [94.0, 105.0], "gt_se_norm": [-0.3671029806137085, -0.29304057359695435], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002767.pt"}
{"idx": 2769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vI_B7dtF7Q0.mp4", "duration": 297.04675, "prompt_text": "Please find the visual event described by the sentence 'Show me when the cow says it was the first cow on the moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 195.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the cow says it was the first cow on the moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 195.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [195.0, 201.0], "base_se_norm": [0.3129246234893799, 0.35332226753234863], "gt_se_raw": [191.0, 195.0], "gt_se_norm": [0.2859928607940674, 0.3129246234893799], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002769.pt"}
{"idx": 2771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6fan6ggvh4U.mp4", "duration": 178.2781, "prompt_text": "Please find the visual event described by the sentence 'When can we see the three siblings sleeping at their desks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the three siblings sleeping at their desks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.697102427482605, -0.6185734272003174], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [-0.697102427482605, -0.6185734272003174], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002771.pt"}
{"idx": 2773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6fan6ggvh4U.mp4", "duration": 178.2781, "prompt_text": "Please find the visual event described by the sentence 'At what point does the character in brown pants throw a crumpled paper ball into a trash can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the character in brown pants throw a crumpled paper ball into a trash can?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 84.0], "base_se_norm": [-0.08008897304534912, -0.05765211582183838], "gt_se_raw": [82.0, 83.0], "gt_se_norm": [-0.08008897304534912, -0.06887054443359375], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002773.pt"}
{"idx": 2775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6fan6ggvh4U.mp4", "duration": 178.2781, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the male character in brown pants points at a portrait of Albert Einstein.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the male character in brown pants points at a portrait of Albert Einstein.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 120.0], "base_se_norm": [0.3013375997543335, 0.346211314201355], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.31255602836608887, 0.346211314201355], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002775.pt"}
{"idx": 2777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUVwR0rw5fk.mp4", "duration": 6.881874999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a 'PAWN SHOP' sign is visible behind the main character.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a 'PAWN SHOP' sign is visible behind the main character.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, 0.7437108755111694], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, 0.7437108755111694], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002777.pt"}
{"idx": 2779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hUVwR0rw5fk.mp4", "duration": 6.881874999999999, "prompt_text": "Please find the visual event described by the sentence 'At what time is a crowd of people looking up at a man on barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a crowd of people looking up at a man on barrels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, 0.7437108755111694], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, 0.7437108755111694], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002779.pt"}
{"idx": 2781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/63d0vV0kk_Y.mp4", "duration": 111.74496666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a website with the title 'Loui Lord Nelson, Ph.D.' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a website with the title 'Loui Lord Nelson, Ph.D.' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 35.0], "base_se_norm": [-0.5525525808334351, -0.37357354164123535], "gt_se_raw": [25.0, 31.0], "gt_se_norm": [-0.5525525808334351, -0.44516515731811523], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002781.pt"}
{"idx": 2783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/63d0vV0kk_Y.mp4", "duration": 111.74496666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are three teddy bears shown holding a picture of a skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are three teddy bears shown holding a picture of a skull?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 72.0], "base_se_norm": [0.23495495319366455, 0.28864872455596924], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [0.21705710887908936, 0.28864872455596924], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002783.pt"}
{"idx": 2785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/63d0vV0kk_Y.mp4", "duration": 111.74496666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the speaker talks about interviewing educators while a photo of two women is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the speaker talks about interviewing educators while a photo of two women is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 59.0], "base_se_norm": [-0.05141139030456543, 0.055976033210754395], "gt_se_raw": [53.0, 60.0], "gt_se_norm": [-0.05141139030456543, 0.07387387752532959], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002785.pt"}
{"idx": 2787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WP_L1I4lmFE.mp4", "duration": 93.32, "prompt_text": "Please find the visual event described by the sentence 'At what point do the two clownfish realize the blue fish is sleep-swimming away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the two clownfish realize the blue fish is sleep-swimming away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 46.0], "base_se_norm": [-0.24989283084869385, -0.0141448974609375], "gt_se_raw": [30.0, 36.0], "gt_se_norm": [-0.3570510149002075, -0.22846120595932007], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002787.pt"}
{"idx": 2789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WP_L1I4lmFE.mp4", "duration": 93.32, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where many small, colorful fish are gathered around the blue fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where many small, colorful fish are gathered around the blue fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 66.0], "base_se_norm": [0.2430347204208374, 0.4144878387451172], "gt_se_raw": [58.0, 66.0], "gt_se_norm": [0.2430347204208374, 0.4144878387451172], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002789.pt"}
{"idx": 2791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ipnGPeRIy2k.mp4", "duration": 228.7285, "prompt_text": "Please find the visual event described by the sentence 'When can we see the question 'What is twerking?' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the question 'What is twerking?' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 71.0], "base_se_norm": [-0.3966646194458008, -0.37917661666870117], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.3966646194458008, -0.37917661666870117], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002791.pt"}
{"idx": 2793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ipnGPeRIy2k.mp4", "duration": 228.7285, "prompt_text": "Please find the visual event described by the sentence 'At what point does the host, Steve Harvey, laugh out loud at the ladies' response?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 203.0 - 212.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the host, Steve Harvey, laugh out loud at the ladies' response?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 203.0 - 212.0 seconds.<|im_end|>\n", "base_se_raw": [203.0, 212.0], "base_se_norm": [0.775030255317688, 0.8537261486053467], "gt_se_raw": [202.0, 206.0], "gt_se_norm": [0.7662862539291382, 0.8012621402740479], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002793.pt"}
{"idx": 2795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ipnGPeRIy2k.mp4", "duration": 228.7285, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the lady in the green jacket expresses her frustration with the new iPhone operating system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the lady in the green jacket expresses her frustration with the new iPhone operating system.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 126.0], "base_se_norm": [-0.16057682037353516, 0.10174286365509033], "gt_se_raw": [100.0, 112.0], "gt_se_norm": [-0.1256008744239807, -0.020672976970672607], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002795.pt"}
{"idx": 2797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bvylqPgr-bQ.mp4", "duration": 180.14663333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a hand writes 'DISABILITY DISCRIMINATION ACT'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a hand writes 'DISABILITY DISCRIMINATION ACT'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8556731343269348, -0.8223669528961182], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.8556731343269348, -0.8223669528961182], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002797.pt"}
{"idx": 2799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bvylqPgr-bQ.mp4", "duration": 180.14663333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show a classroom setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 133.0], "base_se_norm": [0.2434314489364624, 0.4765748977661133], "gt_se_raw": [112.0, 114.0], "gt_se_norm": [0.2434314489364624, 0.26563560962677], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002799.pt"}
{"idx": 2801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pX7dCnUMpbk.mp4", "duration": 110.235125, "prompt_text": "Please find the visual event described by the sentence 'Show me the man introducing himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man introducing himself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9818569421768188, -0.8911417722702026], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.8911417722702026], "start_token_span": [61, 64], "end_token_span": [66, 69], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002801.pt"}
{"idx": 2803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pX7dCnUMpbk.mp4", "duration": 110.235125, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an older man puts his arm around a young man's shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an older man puts his arm around a young man's shoulder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 37.0], "base_se_norm": [-0.5827101469039917, -0.32870757579803467], "gt_se_raw": [23.0, 37.0], "gt_se_norm": [-0.5827101469039917, -0.32870757579803467], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002803.pt"}
{"idx": 2805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pX7dCnUMpbk.mp4", "duration": 110.235125, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the young man is riding a horse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the young man is riding a horse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 97.0], "base_se_norm": [0.5603013038635254, 0.759874701499939], "gt_se_raw": [85.0, 98.0], "gt_se_norm": [0.5421582460403442, 0.7780177593231201], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002805.pt"}
{"idx": 2807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZrO_qsIJw6o.mp4", "duration": 182.91624958291627, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man is using a vacuum cleaner on a dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man is using a vacuum cleaner on a dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 76.0], "base_se_norm": [-0.2564902901649475, -0.16901856660842896], "gt_se_raw": [68.0, 77.0], "gt_se_norm": [-0.2564902901649475, -0.15808463096618652], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002807.pt"}
{"idx": 2809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZrO_qsIJw6o.mp4", "duration": 182.91624958291627, "prompt_text": "Please find the visual event described by the sentence 'At what point is a dog seen running on a frozen lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a dog seen running on a frozen lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 114.0], "base_se_norm": [0.08246266841888428, 0.24647212028503418], "gt_se_raw": [99.0, 114.0], "gt_se_norm": [0.08246266841888428, 0.24647212028503418], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002809.pt"}
{"idx": 2811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xLE1rO60uCk.mp4", "duration": 69.83643333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 69.0], "base_se_norm": [-0.9713616371154785, 0.9760459661483765], "gt_se_raw": [0.0, 68.0], "gt_se_norm": [-1.0, 0.947407603263855], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002811.pt"}
{"idx": 2813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xLE1rO60uCk.mp4", "duration": 69.83643333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker say the words 'real time machines'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker say the words 'real time machines'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [0.0023421049118041992, 0.08825719356536865], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.08357292413711548, -0.02629619836807251], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002813.pt"}
{"idx": 2815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xLE1rO60uCk.mp4", "duration": 69.83643333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man wish the audience 'clear skies' and 'good tailwinds'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man wish the audience 'clear skies' and 'good tailwinds'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 60.0], "base_se_norm": [0.6037473678588867, 0.7183008193969727], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [0.8042159080505371, 0.8614925146102905], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002815.pt"}
{"idx": 2817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-9WnLz26Bwc.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the moose is swimming across the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the moose is swimming across the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 11.0], "base_se_norm": [-0.9265140295028687, -0.7305514216423035], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.902018666267395, -0.8040373921394348], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002817.pt"}
{"idx": 2819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-9WnLz26Bwc.mp4", "duration": 81.64823333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the moose is eating leaves from a bush on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the moose is eating leaves from a bush on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 58.0], "base_se_norm": [-0.3876168727874756, 0.4207289218902588], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.5345888137817383, -0.41211217641830444], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002819.pt"}
{"idx": 2821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h56Cr2ho1Y8.mp4", "duration": 186.3528333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man and a woman doing jump squats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man and a woman doing jump squats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.420454204082489, -0.3238632082939148], "gt_se_raw": [54.0, 62.0], "gt_se_norm": [-0.420454204082489, -0.33459556102752686], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002821.pt"}
{"idx": 2823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h56Cr2ho1Y8.mp4", "duration": 186.3528333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a woman reading a book on a bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a woman reading a book on a bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [-0.6887624263763428, -0.6565654277801514], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.6887624263763428, -0.6672977805137634], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002823.pt"}
{"idx": 2825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h56Cr2ho1Y8.mp4", "duration": 186.3528333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are they performing burpees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are they performing burpees?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 135.0], "base_se_norm": [-0.27020156383514404, 0.4488644599914551], "gt_se_raw": [109.0, 135.0], "gt_se_norm": [0.16982388496398926, 0.4488644599914551], "start_token_span": [63, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002825.pt"}
{"idx": 2827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K-RVlLR7wMw.mp4", "duration": 29.9299, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the woman in the gray shirt adjusts the girl's white hair clip.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the woman in the gray shirt adjusts the girl's white hair clip.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.6658859252929688, -0.3317719101905823], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.5990631580352783, -0.3985946774482727], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002827.pt"}
{"idx": 2829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K-RVlLR7wMw.mp4", "duration": 29.9299, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher standing up point at the cartoon dog on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher standing up point at the cartoon dog on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.13130342960357666], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.47010183334350586, 0.6037474870681763], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002829.pt"}
{"idx": 2831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TvMAUk3Etjw.mp4", "duration": 187.4, "prompt_text": "Please find the visual event described by the sentence 'When does a player stack up a lot of TNT blocks and blow them up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a player stack up a lot of TNT blocks and blow them up?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9466382265090942, -0.8932763934135437], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.9359658360481262, -0.9039487838745117], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002831.pt"}
{"idx": 2833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TvMAUk3Etjw.mp4", "duration": 187.4, "prompt_text": "Please find the visual event described by the sentence 'When is the player with the pumpkin head knocked off the map by another player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the player with the pumpkin head knocked off the map by another player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8612593412399292, -0.8292422294616699], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.5197438597679138, -0.49839913845062256], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002833.pt"}
{"idx": 2835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TvMAUk3Etjw.mp4", "duration": 187.4, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone making a TNT cannon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone making a TNT cannon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.9466382265090942, -0.9039487838745117], "gt_se_raw": [137.0, 142.0], "gt_se_norm": [0.4621131420135498, 0.5154749155044556], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002835.pt"}
{"idx": 2837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dNJdJIwCF_Y.mp4", "duration": 100.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a baseball is being cut into small dice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a baseball is being cut into small dice.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 36.0], "base_se_norm": [-0.5239669680595398, -0.2859504222869873], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.4644628167152405, -0.2859504222869873], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002837.pt"}
{"idx": 2839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dNJdJIwCF_Y.mp4", "duration": 100.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a chess piece being used as a salt shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a chess piece being used as a salt shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [0.4082643985748291, 0.4876033067703247], "gt_se_raw": [72.0, 75.0], "gt_se_norm": [0.4280991554260254, 0.4876033067703247], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002839.pt"}
{"idx": 2841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LdmGQg6T29A.mp4", "duration": 22.389033333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in purple smile happily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in purple smile happily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.6426821947097778, -0.4640233516693115], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7320116758346558, -0.4640233516693115], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002841.pt"}
{"idx": 2843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LdmGQg6T29A.mp4", "duration": 22.389033333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in black hold the girl's hands and lean in close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in black hold the girl's hands and lean in close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.4640233516693115], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.5533527731895447], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002843.pt"}
{"idx": 2845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LdmGQg6T29A.mp4", "duration": 22.389033333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the girl wearing a nasal cannula most clearly visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl wearing a nasal cannula most clearly visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, 0.9652477502822876], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9106705784797668, -0.6426821947097778], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002845.pt"}
{"idx": 2847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tqpPFT-F-bs.mp4", "duration": 127.5274, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the text 'ONE SLAP WILL BRING THEM TOGETHER' appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the text 'ONE SLAP WILL BRING THEM TOGETHER' appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.49814707040786743, -0.45109832286834717], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [-0.48246413469314575, -0.45109832286834717], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002847.pt"}
{"idx": 2849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tqpPFT-F-bs.mp4", "duration": 127.5274, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a striped sweater slaps another man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a striped sweater slaps another man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 100.0], "base_se_norm": [0.5055588483810425, 0.5682904720306396], "gt_se_raw": [102.0, 103.0], "gt_se_norm": [0.5996562242507935, 0.6153391599655151], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002849.pt"}
{"idx": 2851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9cW0Hlj_P-A.mp4", "duration": 189.68285828984344, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where hands are holding up letters that spell 'Respect'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where hands are holding up letters that spell 'Respect'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 29.0], "base_se_norm": [-0.7153142690658569, -0.6942265033721924], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.7153142690658569, -0.6836825609207153], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002851.pt"}
{"idx": 2853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9cW0Hlj_P-A.mp4", "duration": 189.68285828984344, "prompt_text": "Please find the visual event described by the sentence 'When does the character ask, 'Is he retarded?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character ask, 'Is he retarded?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 63.0], "base_se_norm": [-0.3568211793899536, -0.3357333540916443], "gt_se_raw": [61.0, 62.0], "gt_se_norm": [-0.3568211793899536, -0.34627723693847656], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002853.pt"}
{"idx": 2855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9cW0Hlj_P-A.mp4", "duration": 189.68285828984344, "prompt_text": "Please find the visual event described by the sentence 'Show me the slide about how to improve the language used.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the slide about how to improve the language used.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 189.0], "base_se_norm": [0.7819216251373291, 0.9927999973297119], "gt_se_raw": [169.0, 189.0], "gt_se_norm": [0.7819216251373291, 0.9927999973297119], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002855.pt"}
{"idx": 2857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3heXk6Oj6hU.mp4", "duration": 75.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a woman's hand enters the frame and points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a woman's hand enters the frame and points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.6815568208694458, -0.5754091143608093], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.6550198793411255, -0.5754091143608093], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002857.pt"}
{"idx": 2859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3heXk6Oj6hU.mp4", "duration": 75.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the woman's hand guiding the person's hand to write?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman's hand guiding the person's hand to write?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 63.0], "base_se_norm": [0.008403301239013672, 0.6718266010284424], "gt_se_raw": [46.0, 60.0], "gt_se_norm": [0.22069871425628662, 0.5922157764434814], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002859.pt"}
{"idx": 2861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QL3H7CUJMDU.mp4", "duration": 160.84, "prompt_text": "Please find the visual event described by the sentence 'When does the fat man confess his love to the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the fat man confess his love to the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 47.0], "base_se_norm": [-0.4777418375015259, -0.4155682325363159], "gt_se_raw": [43.0, 44.0], "gt_se_norm": [-0.4653071165084839, -0.4528723955154419], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002861.pt"}
{"idx": 2863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QL3H7CUJMDU.mp4", "duration": 160.84, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man offer flowers to the police officer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man offer flowers to the police officer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 103.0], "base_se_norm": [0.24347186088562012, 0.28077590465545654], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.25590646266937256, 0.28077590465545654], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002863.pt"}
{"idx": 2865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QL3H7CUJMDU.mp4", "duration": 160.84, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man with a mustache has his son, represented by a ball, kicked into the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man with a mustache has his son, represented by a ball, kicked into the sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [-0.30365580320358276, -0.2663516402244568], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [-0.2787863612174988, -0.2539169192314148], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002865.pt"}
{"idx": 2867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ISQfQ6XmwQo.mp4", "duration": 31.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a woman in a black shirt looking through files in a library.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a woman in a black shirt looking through files in a library.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [-0.05163329839706421, 0.07481563091278076], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.05163329839706421, 0.07481563091278076], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002867.pt"}
{"idx": 2869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ISQfQ6XmwQo.mp4", "duration": 31.633333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a male student is using a stethoscope on a medical dummy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a male student is using a stethoscope on a medical dummy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [0.32771337032318115, 0.5173866748809814], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [0.32771337032318115, 0.45416224002838135], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002869.pt"}
{"idx": 2871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/75Qg6ldZNbA.mp4", "duration": 198.03116666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the completed blue paper boat at the beginning of the tutorial?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the completed blue paper boat at the beginning of the tutorial?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.8990057706832886, -0.8485087156295776], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.8889063596725464, -0.8384093046188354], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002871.pt"}
{"idx": 2873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/75Qg6ldZNbA.mp4", "duration": 198.03116666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point are the bottom rectangular flaps folded upwards on both sides?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the bottom rectangular flaps folded upwards on both sides?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 142.0], "base_se_norm": [0.24222862720489502, 0.4341176748275757], "gt_se_raw": [77.0, 94.0], "gt_se_norm": [-0.22234463691711426, -0.05065453052520752], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002873.pt"}
{"idx": 2875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/75Qg6ldZNbA.mp4", "duration": 198.03116666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the final step where the sides are pulled apart to form the boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the final step where the sides are pulled apart to form the boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 183.0], "base_se_norm": [0.636106014251709, 0.848193883895874], "gt_se_raw": [162.0, 173.0], "gt_se_norm": [0.636106014251709, 0.7471996545791626], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002875.pt"}
{"idx": 2877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rbNB0jqMv7s.mp4", "duration": 54.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the instructor in the red suit guides the first person out of the wind tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the instructor in the red suit guides the first person out of the wind tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.04294478893280029, 0.17791414260864258], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.04294478893280029, 0.06748473644256592], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002877.pt"}
{"idx": 2879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rbNB0jqMv7s.mp4", "duration": 54.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the second person in the blue suit enters the wind tunnel through the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the second person in the blue suit enters the wind tunnel through the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [0.8036810159683228, 0.9141104221343994], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.8036810159683228, 0.9141104221343994], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002879.pt"}
{"idx": 2881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/StAF3NSro-w.mp4", "duration": 169.3, "prompt_text": "Please find the visual event described by the sentence 'When does the person choose the yellow ball and put it in the wooden box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person choose the yellow ball and put it in the wooden box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 36.0], "base_se_norm": [-0.704666256904602, -0.5747194290161133], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.6219727993011475, -0.5865328311920166], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002881.pt"}
{"idx": 2883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/StAF3NSro-w.mp4", "duration": 169.3, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy rejects the choices, and the man signs 'finished'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy rejects the choices, and the man signs 'finished'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 141.0], "base_se_norm": [0.5593620538711548, 0.6656821966171265], "gt_se_raw": [95.0, 99.0], "gt_se_norm": [0.1222681999206543, 0.16952157020568848], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002883.pt"}
{"idx": 2885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/StAF3NSro-w.mp4", "duration": 169.3, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the boy picks the black square object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the boy picks the black square object.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 114.0], "base_se_norm": [0.27584171295166016, 0.3467217683792114], "gt_se_raw": [117.0, 124.0], "gt_se_norm": [0.38216185569763184, 0.4648553133010864], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002885.pt"}
{"idx": 2887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvfxpJn7VRM.mp4", "duration": 53.59520833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man walks in from the left holding a blue sledgehammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man walks in from the left holding a blue sledgehammer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.8134161233901978, -0.6641490459442139], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.8134161233901978, -0.6641490459442139], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002887.pt"}
{"idx": 2889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvfxpJn7VRM.mp4", "duration": 53.59520833333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man swings the hammer and smashes the pie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man swings the hammer and smashes the pie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.3656148910522461, -0.25366461277008057], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.3656148910522461, -0.29098135232925415], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002889.pt"}
{"idx": 2891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WnW-fl0mNwo.mp4", "duration": 79.81306666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the pitcher throw the final pitch of the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the pitcher throw the final pitch of the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.9248243570327759, -0.8747072219848633], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9248243570327759, -0.8747072219848633], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002891.pt"}
{"idx": 2893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WnW-fl0mNwo.mp4", "duration": 79.81306666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the losing team's players looking dejected in the dugout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the losing team's players looking dejected in the dugout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 74.0], "base_se_norm": [0.3281034231185913, 0.8543330430984497], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [0.3281034231185913, 0.40327906608581543], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002893.pt"}
{"idx": 2895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WnW-fl0mNwo.mp4", "duration": 79.81306666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the winning team forms a massive celebratory huddle on the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the winning team forms a massive celebratory huddle on the field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 22.0], "base_se_norm": [-0.7494144439697266, -0.4487118124961853], "gt_se_raw": [9.0, 22.0], "gt_se_norm": [-0.7744730114936829, -0.4487118124961853], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002895.pt"}
{"idx": 2897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34zJjqZoX4E.mp4", "duration": 101.39295833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man place his briefcase on the floor and put his hands on his hips?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man place his briefcase on the floor and put his hands on his hips?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.4674186706542969, -0.36879247426986694], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.4674186706542969, -0.36879247426986694], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002897.pt"}
{"idx": 2899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34zJjqZoX4E.mp4", "duration": 101.39295833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man calls out for his family but gets no response.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man calls out for his family but gets no response.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 78.0], "base_se_norm": [0.14406371116638184, 0.538568377494812], "gt_se_raw": [53.0, 66.0], "gt_se_norm": [0.04543745517730713, 0.3018655776977539], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002899.pt"}
{"idx": 2901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mG4Y2Snygfk.mp4", "duration": 59.02563333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the drawing of two stick figures hugging?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the drawing of two stick figures hugging?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.08514326810836792, 0.016507506370544434], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.08514326810836792, 0.016507506370544434], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002901.pt"}
{"idx": 2903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mG4Y2Snygfk.mp4", "duration": 59.02563333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the phrase 'Become aware. Don't discriminate.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the phrase 'Become aware. Don't discriminate.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 49.0], "base_se_norm": [0.45699405670166016, 0.6602956056594849], "gt_se_raw": [43.0, 49.0], "gt_se_norm": [0.45699405670166016, 0.6602956056594849], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002903.pt"}
{"idx": 2905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mG4Y2Snygfk.mp4", "duration": 59.02563333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when negative words like 'Helpless' and 'victim' are shown on the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when negative words like 'Helpless' and 'victim' are shown on the page.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.8983492255210876, -0.7289313077926636], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.8983492255210876, -0.7289313077926636], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002905.pt"}
{"idx": 2907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eXDSa-7VWpI.mp4", "duration": 117.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with the large red beard embrace the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with the large red beard embrace the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 26.0], "base_se_norm": [-0.6427558660507202, -0.5576977729797363], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.6257442235946655, -0.5917209982872009], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002907.pt"}
{"idx": 2909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eXDSa-7VWpI.mp4", "duration": 117.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the blonde girl kisses the boy on the cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the blonde girl kisses the boy on the cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.08137226104736328, -0.013325750827789307], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [-0.0643606185913086, -0.030337393283843994], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002909.pt"}
{"idx": 2911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sEqSIwf-0Lg.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'When do we see someone adding coffee grounds to the filter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see someone adding coffee grounds to the filter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.6001999378204346, -0.5252373814582825], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.6001999378204346, -0.5252373814582825], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002911.pt"}
{"idx": 2913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sEqSIwf-0Lg.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the dripper is being removed from the glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the dripper is being removed from the glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 49.0], "base_se_norm": [0.17441272735595703, 0.2243877649307251], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.17441272735595703, 0.2243877649307251], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002913.pt"}
{"idx": 2915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sEqSIwf-0Lg.mp4", "duration": 80.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where milk is added to the black coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where milk is added to the black coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [0.2993502616882324, 0.3743128776550293], "gt_se_raw": [52.0, 55.0], "gt_se_norm": [0.2993502616882324, 0.3743128776550293], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002915.pt"}
{"idx": 2917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KY-4VyUY8hk.mp4", "duration": 151.88506666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an egg is added to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an egg is added to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 58.0], "base_se_norm": [-0.5259573459625244, -0.2362646460533142], "gt_se_raw": [36.0, 57.0], "gt_se_norm": [-0.5259573459625244, -0.2494325041770935], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002917.pt"}
{"idx": 2919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KY-4VyUY8hk.mp4", "duration": 151.88506666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when chocolate chips are mixed in.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when chocolate chips are mixed in.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 89.0], "base_se_norm": [0.013924479484558105, 0.17193877696990967], "gt_se_raw": [81.0, 89.0], "gt_se_norm": [0.06659591197967529, 0.17193877696990967], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002919.pt"}
{"idx": 2921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KauXf6nihPY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When are the paper ABCD cards being folded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the paper ABCD cards being folded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.7379669547080994], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8034752011299133], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002921.pt"}
{"idx": 2923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KauXf6nihPY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When is the user searching for the app in the App Store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the user searching for the app in the App Store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.14839261770248413, 0.1136404275894165], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.14839261770248413, 0.1136404275894165], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002923.pt"}
{"idx": 2925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KauXf6nihPY.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point do the students hold up their phones displaying the letter 'B'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the students hold up their phones displaying the letter 'B'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [0.6377065181732178, 0.7687230110168457], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [0.6377065181732178, 0.7687230110168457], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002925.pt"}
{"idx": 2927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n2Y6in_CZ6A.mp4", "duration": 173.0397063730397, "prompt_text": "Please find the visual event described by the sentence 'At what point is a diver seen cleaning the acrylic in the kelp forest exhibit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a diver seen cleaning the acrylic in the kelp forest exhibit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 90.0], "base_se_norm": [-0.029124557971954346, 0.04022371768951416], "gt_se_raw": [84.0, 90.0], "gt_se_norm": [-0.029124557971954346, 0.04022371768951416], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002927.pt"}
{"idx": 2929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n2Y6in_CZ6A.mp4", "duration": 173.0397063730397, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a corn snake is climbing on a wooden structure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a corn snake is climbing on a wooden structure.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 137.0], "base_se_norm": [0.41008102893829346, 0.5834516286849976], "gt_se_raw": [122.0, 137.0], "gt_se_norm": [0.41008102893829346, 0.5834516286849976], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002929.pt"}
{"idx": 2931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F85cuRuQjLE.mp4", "duration": 193.72, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man injecting the cow with medicine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man injecting the cow with medicine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 90.0], "base_se_norm": [-0.15341728925704956, -0.07082384824752808], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [-0.14309310913085938, -0.09147220849990845], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002931.pt"}
{"idx": 2933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F85cuRuQjLE.mp4", "duration": 193.72, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the cow finally manages to stand up on its feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the cow finally manages to stand up on its feet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 168.0], "base_se_norm": [0.6931654214859009, 0.7344621419906616], "gt_se_raw": [163.0, 167.0], "gt_se_norm": [0.6828411817550659, 0.7241379022598267], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002933.pt"}
{"idx": 2935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F85cuRuQjLE.mp4", "duration": 193.72, "prompt_text": "Please find the visual event described by the sentence 'When can the sound of a siren be heard approaching?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can the sound of a siren be heard approaching?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 52.0], "base_se_norm": [-0.6180053949356079, -0.4631426930427551], "gt_se_raw": [17.0, 37.0], "gt_se_norm": [-0.8244889378547668, -0.6180053949356079], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002935.pt"}
{"idx": 2937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3CJl_S7uzqU.mp4", "duration": 124.49103333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man make a 'stop' gesture with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man make a 'stop' gesture with his hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [-0.7590187788009644, -0.7268879413604736], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.7590187788009644, -0.7268879413604736], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002937.pt"}
{"idx": 2939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3CJl_S7uzqU.mp4", "duration": 124.49103333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when they are playing the Connect Four board game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when they are playing the Connect Four board game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 49.0], "base_se_norm": [-0.40557968616485596, -0.2127947211265564], "gt_se_raw": [37.0, 49.0], "gt_se_norm": [-0.40557968616485596, -0.2127947211265564], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002939.pt"}
{"idx": 2941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zOVoL1-ZaQQ.mp4", "duration": 145.0, "prompt_text": "Please find the visual event described by the sentence 'When is the man with the white beard and blue shirt speaking for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man with the white beard and blue shirt speaking for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 17.0], "base_se_norm": [-0.9448275566101074, -0.7655172348022461], "gt_se_raw": [4.0, 17.0], "gt_se_norm": [-0.9448275566101074, -0.7655172348022461], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002941.pt"}
{"idx": 2943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zOVoL1-ZaQQ.mp4", "duration": 145.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman in the red top using sign language while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman in the red top using sign language while speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 81.0], "base_se_norm": [-0.10344827175140381, 0.11724138259887695], "gt_se_raw": [65.0, 82.0], "gt_se_norm": [-0.10344827175140381, 0.1310344934463501], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002943.pt"}
{"idx": 2945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zOVoL1-ZaQQ.mp4", "duration": 145.0, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man in the blue shirt with a white t-shirt underneath smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man in the blue shirt with a white t-shirt underneath smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.682758629322052, -0.613793134689331], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.6275862455368042, -0.6000000238418579], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002945.pt"}
{"idx": 2947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/exKximzPvAs.mp4", "duration": 101.51808333333332, "prompt_text": "Please find the visual event described by the sentence 'Where is the packet of instant hot cocoa mix shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where is the packet of instant hot cocoa mix shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.5468787550926208, -0.4680750370025635], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.5468787550926208, -0.4483741521835327], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002947.pt"}
{"idx": 2949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/exKximzPvAs.mp4", "duration": 101.51808333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the book cover with a drawing of a s'mores sale cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the book cover with a drawing of a s'mores sale cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 84.0], "base_se_norm": [0.5957748889923096, 0.6548776626586914], "gt_se_raw": [81.0, 84.0], "gt_se_norm": [0.5957748889923096, 0.6548776626586914], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002949.pt"}
{"idx": 2951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y-Z7LObUlwA.mp4", "duration": 155.3552, "prompt_text": "Please find the visual event described by the sentence 'When is the tour guide introducing himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the tour guide introducing himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 22.0], "base_se_norm": [-0.961378812789917, -0.716778039932251], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.794020414352417, -0.7682729363441467], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002951.pt"}
{"idx": 2953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y-Z7LObUlwA.mp4", "duration": 155.3552, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman with blue hair lead the tour down the hallway to the testing space?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman with blue hair lead the tour down the hallway to the testing space?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 63.0], "base_se_norm": [-0.3949349522590637, -0.1889553666114807], "gt_se_raw": [56.0, 62.0], "gt_se_norm": [-0.2790713906288147, -0.20182907581329346], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002953.pt"}
{"idx": 2955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y-Z7LObUlwA.mp4", "duration": 155.3552, "prompt_text": "Please find the visual event described by the sentence 'When is the man named Dan shown signing his name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man named Dan shown signing his name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 105.0], "base_se_norm": [0.08139288425445557, 0.35174107551574707], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.09426665306091309, 0.13288772106170654], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002955.pt"}
{"idx": 2957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K7bNhYr3ves.mp4", "duration": 84.61786666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the flashback scene where a single red fish egg is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the flashback scene where a single red fish egg is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.48001527786254883, -0.3854725956916809], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.48001527786254883, -0.36183691024780273], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002957.pt"}
{"idx": 2959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K7bNhYr3ves.mp4", "duration": 84.61786666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does Marlin tell Nemo that he met a 150-year-old sea turtle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Marlin tell Nemo that he met a 150-year-old sea turtle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 60.0], "base_se_norm": [0.25269055366516113, 0.418140172958374], "gt_se_raw": [55.0, 62.0], "gt_se_norm": [0.2999619245529175, 0.46541154384613037], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002959.pt"}
{"idx": 2961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpNiqsM81K0.mp4", "duration": 41.37466666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the hands fully interlocked and stable before any major untangling movement begins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the hands fully interlocked and stable before any major untangling movement begins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.37159615755081177], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.8066449761390686, -0.5649511814117432], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002961.pt"}
{"idx": 2963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpNiqsM81K0.mp4", "duration": 41.37466666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the person on the right raises the other person's thumb up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the person on the right raises the other person's thumb up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.46827369928359985, -0.3232573866844177], "gt_se_raw": [24.0, 25.0], "gt_se_norm": [0.16013014316558838, 0.20846891403198242], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002963.pt"}
{"idx": 2965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FpNiqsM81K0.mp4", "duration": 41.37466666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the final step of the trick, where one person points with their index finger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the final step of the trick, where one person points with their index finger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [0.6918565034866333, 0.981889009475708], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [0.8368726968765259, 0.933550238609314], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002965.pt"}
{"idx": 2967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/igOTq9Z7bm0.mp4", "duration": 86.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the child in the blue shirt is lying on the floor and the man is holding his stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the child in the blue shirt is lying on the floor and the man is holding his stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [-0.7926068305969238, -0.6543446779251099], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.7926068305969238, -0.6543446779251099], "start_token_span": [77, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002967.pt"}
{"idx": 2969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/igOTq9Z7bm0.mp4", "duration": 86.79166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man blows a raspberry on the child's belly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man blows a raspberry on the child's belly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.4008641242980957, -0.3086893558502197], "gt_se_raw": [27.0, 40.0], "gt_se_norm": [-0.3778204321861267, -0.07825249433517456], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002969.pt"}
{"idx": 2971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-MuOaopJY64.mp4", "duration": 9.509500000000001, "prompt_text": "Please find the visual event described by the sentence 'When are a man and a child seen walking down a long corridor from behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are a man and a child seen walking down a long corridor from behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.8928440809249878], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, 0.8928440809249878], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002971.pt"}
{"idx": 2973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-MuOaopJY64.mp4", "duration": 9.509500000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment they are walking calmly together after the child slows down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment they are walking calmly together after the child slows down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.158735990524292, 0.682528018951416], "gt_se_raw": [3.0, 9.0], "gt_se_norm": [-0.369051992893219, 0.8928440809249878], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002973.pt"}
{"idx": 2975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-MuOaopJY64.mp4", "duration": 9.509500000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point do they pass the second window on the right wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they pass the second window on the right wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 7.0], "base_se_norm": [0.051580071449279785, 0.4722120761871338], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [0.051580071449279785, 0.682528018951416], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002975.pt"}
{"idx": 2977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-_kGBuX1Qy0.mp4", "duration": 11.0, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where the boy in the blue striped shirt is holding a red object and looking at the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where the boy in the blue striped shirt is holding a red object and looking at the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.8181818723678589], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.45454543828964233, 0.27272725105285645], "start_token_span": [82, 85], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002977.pt"}
{"idx": 2979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-_kGBuX1Qy0.mp4", "duration": 11.0, "prompt_text": "Please find the visual event described by the sentence 'At what time is the girl in the wheelchair seen with a white bib on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the girl in the wheelchair seen with a white bib on?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.8181818723678589], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, 0.8181818723678589], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002979.pt"}
{"idx": 2981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O9T25SMyz3A.mp4", "duration": 171.80496666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man collapse after experiencing discomfort?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man collapse after experiencing discomfort?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5925612449645996, -0.5576379299163818], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.5925612449645996, -0.5692790150642395], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002981.pt"}
{"idx": 2983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O9T25SMyz3A.mp4", "duration": 171.80496666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the Automated External Defibrillator (AED) is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the Automated External Defibrillator (AED) is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 55.0], "base_se_norm": [-0.41794461011886597, -0.35973912477493286], "gt_se_raw": [50.0, 55.0], "gt_se_norm": [-0.41794461011886597, -0.35973912477493286], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002983.pt"}
{"idx": 2985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O9T25SMyz3A.mp4", "duration": 171.80496666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man in a red shirt demonstrates chest compressions on a mannequin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man in a red shirt demonstrates chest compressions on a mannequin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 129.0], "base_se_norm": [0.07098186016082764, 0.5017027854919434], "gt_se_raw": [93.0, 129.0], "gt_se_norm": [0.08262300491333008, 0.5017027854919434], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002985.pt"}
{"idx": 2987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkDMIKoZDOs.mp4", "duration": 30.466666666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the athlete is warming up by stretching their arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the athlete is warming up by stretching their arms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.4091903567314148, -0.2122538685798645], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.4748358726501465, -0.2122538685798645], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002987.pt"}
{"idx": 2989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bkDMIKoZDOs.mp4", "duration": 30.466666666666665, "prompt_text": "Please find the visual event described by the sentence 'Can you identify the time range when the 'SNACK BAR' sign is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you identify the time range when the 'SNACK BAR' sign is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [0.24726474285125732, 0.3785557746887207], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [0.1816192865371704, 0.4442012310028076], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002989.pt"}
{"idx": 2991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WH50zXiSVPg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is much higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9030095934867859, -0.4180576801300049], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9030095934867859, -0.4180576801300049], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002991.pt"}
{"idx": 2993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WH50zXiSVPg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Show me the bar chart that compares employment rates between two different groups over two years.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the bar chart that compares employment rates between two different groups over two years.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002993.pt"}
{"idx": 2995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WH50zXiSVPg.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'Find the period showing the 41.1% employment rate for disabled Americans in 2010.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the period showing the 41.1% employment rate for disabled Americans in 2010.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [83, 86], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002995.pt"}
{"idx": 2997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NQQH5TfyNKc.mp4", "duration": 58.25825825825826, "prompt_text": "Please find the visual event described by the sentence 'Show the screen recording of the Panopto software interface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the screen recording of the Panopto software interface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [-0.2447422742843628, -0.17608249187469482], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.2790721654891968, -0.2104123830795288], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002997.pt"}
{"idx": 2999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NQQH5TfyNKc.mp4", "duration": 58.25825825825826, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a hand touches the screen to point at the 'summary' section.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a hand touches the screen to point at the 'summary' section.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 46.0], "base_se_norm": [0.47618556022644043, 0.5791752338409424], "gt_se_raw": [44.0, 46.0], "gt_se_norm": [0.5105154514312744, 0.5791752338409424], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00002999.pt"}
{"idx": 3001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwBxO3289cE.mp4", "duration": 132.93293293293294, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a woman wearing a red top and a man in a denim jacket standing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a woman wearing a red top and a man in a denim jacket standing together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.5787349939346313, -0.5185542106628418], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.5937801599502563, -0.5335993766784668], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003001.pt"}
{"idx": 3003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwBxO3289cE.mp4", "duration": 132.93293293293294, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a purple sweater is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a purple sweater is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.27783137559890747, -0.20260542631149292], "gt_se_raw": [48.0, 53.0], "gt_se_norm": [-0.27783137559890747, -0.20260542631149292], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003003.pt"}
{"idx": 3005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwBxO3289cE.mp4", "duration": 132.93293293293294, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man with white curly hair sitting at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man with white curly hair sitting at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [0.15847885608673096, 0.2186596393585205], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [0.14343369007110596, 0.2186596393585205], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003005.pt"}
{"idx": 3007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q5-KoPvZDIU.mp4", "duration": 120.08663333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the segment that illustrates the different body parts of a dragonfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment that illustrates the different body parts of a dragonfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 88.0], "base_se_norm": [0.21578896045684814, 0.4656085968017578], "gt_se_raw": [75.0, 87.0], "gt_se_norm": [0.24909818172454834, 0.4489539861679077], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003007.pt"}
{"idx": 3009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q5-KoPvZDIU.mp4", "duration": 120.08663333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see multiple dragonflies resting on a wet log by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see multiple dragonflies resting on a wet log by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 92.0], "base_se_norm": [0.4656085968017578, 0.5322271585464478], "gt_se_raw": [88.0, 92.0], "gt_se_norm": [0.4656085968017578, 0.5322271585464478], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003009.pt"}
{"idx": 3011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8ISzf2pryI.mp4", "duration": 138.0, "prompt_text": "Please find the visual event described by the sentence 'When does the owner confront the golden retriever with the torn treat bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the owner confront the golden retriever with the torn treat bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [-0.6376811265945435, -0.4492753744125366], "gt_se_raw": [25.0, 37.0], "gt_se_norm": [-0.6376811265945435, -0.4637681245803833], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003011.pt"}
{"idx": 3013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8ISzf2pryI.mp4", "duration": 138.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the torn bag of treats first revealed on the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the torn bag of treats first revealed on the carpet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9855072498321533, -0.8115941882133484], "gt_se_raw": [1.0, 10.0], "gt_se_norm": [-0.9855072498321533, -0.8550724983215332], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003013.pt"}
{"idx": 3015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B8ISzf2pryI.mp4", "duration": 138.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the owner declares the second dog is the one who did it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the owner declares the second dog is the one who did it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 108.0], "base_se_norm": [0.49275362491607666, 0.5652173757553101], "gt_se_raw": [80.0, 86.0], "gt_se_norm": [0.15942025184631348, 0.24637675285339355], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003015.pt"}
{"idx": 3017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W2Cxs_p2-oE.mp4", "duration": 127.127, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man in a pink dress and fairy wings is writing in a notepad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man in a pink dress and fairy wings is writing in a notepad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 66.0], "base_se_norm": [-0.46510183811187744, 0.03833174705505371], "gt_se_raw": [34.0, 40.0], "gt_se_norm": [-0.46510183811187744, -0.37070804834365845], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003017.pt"}
{"idx": 3019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W2Cxs_p2-oE.mp4", "duration": 127.127, "prompt_text": "Please find the visual event described by the sentence 'When does the title card for the 'Grumpy Princess Show' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the title card for the 'Grumpy Princess Show' appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.8584092855453491, -0.7482832074165344], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.8584092855453491, -0.7482832074165344], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003019.pt"}
{"idx": 3021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B43fpt5MYxc.mp4", "duration": 151.70833333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a black shirt taking another person's temperature with a forehead thermometer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a black shirt taking another person's temperature with a forehead thermometer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [-0.5122219324111938, -0.44630593061447144], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.5122219324111938, -0.44630593061447144], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003021.pt"}
{"idx": 3023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B43fpt5MYxc.mp4", "duration": 151.70833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are students seen writing at their desks in a socially distanced classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are students seen writing at their desks in a socially distanced classroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 53.0], "base_se_norm": [-0.3803899884223938, -0.3012908101081848], "gt_se_raw": [46.0, 53.0], "gt_se_norm": [-0.39357316493988037, -0.3012908101081848], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003023.pt"}
{"idx": 3025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B43fpt5MYxc.mp4", "duration": 151.70833333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a teacher wearing a face shield is addressing the class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a teacher wearing a face shield is addressing the class.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 68.0], "base_se_norm": [-0.20900851488113403, -0.10354292392730713], "gt_se_raw": [60.0, 68.0], "gt_se_norm": [-0.20900851488113403, -0.10354292392730713], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003025.pt"}
{"idx": 3027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4FE1CLS7i3k.mp4", "duration": 312.8542083333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a hand writing on a piece of light blue paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a hand writing on a piece of light blue paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 42.0], "base_se_norm": [-0.7442898750305176, -0.7315043210983276], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.7442898750305176, -0.7315043210983276], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003027.pt"}
{"idx": 3029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4FE1CLS7i3k.mp4", "duration": 312.8542083333333, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a person writes 'I love School Now!' on a board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 211.0 - 214.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a person writes 'I love School Now!' on a board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 211.0 - 214.0 seconds.<|im_end|>\n", "base_se_raw": [211.0, 214.0], "base_se_norm": [0.348871111869812, 0.3680492639541626], "gt_se_raw": [211.0, 214.0], "gt_se_norm": [0.348871111869812, 0.3680492639541626], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003029.pt"}
{"idx": 3031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bWy7nCH3zmQ.mp4", "duration": 116.34956666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman turn the page for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman turn the page for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8281041979789734, -0.7765355110168457], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.8452937602996826, -0.8109146356582642], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003031.pt"}
{"idx": 3033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bWy7nCH3zmQ.mp4", "duration": 116.34956666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman holds up the book titled 'Today I Feel Silly'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman holds up the book titled 'Today I Feel Silly'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 46.0], "base_se_norm": [-0.31241685152053833, -0.2092793583869934], "gt_se_raw": [40.0, 46.0], "gt_se_norm": [-0.31241685152053833, -0.2092793583869934], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003033.pt"}
{"idx": 3035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bWy7nCH3zmQ.mp4", "duration": 116.34956666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman points to the 'silly, crazy' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman points to the 'silly, crazy' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.6733980178833008, -0.6218292713165283], "gt_se_raw": [108.0, 112.0], "gt_se_norm": [0.8564745187759399, 0.9252328872680664], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003035.pt"}
{"idx": 3037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ua6wauucdOo.mp4", "duration": 151.1092888888889, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where rescue workers are clearing rubble at a disaster site.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where rescue workers are clearing rubble at a disaster site.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 117.0], "base_se_norm": [0.4294290542602539, 0.5485481023788452], "gt_se_raw": [107.0, 117.0], "gt_se_norm": [0.4161936044692993, 0.5485481023788452], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003037.pt"}
{"idx": 3039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ua6wauucdOo.mp4", "duration": 151.1092888888889, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the cricket is shown jumping and landing in slow motion against a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the cricket is shown jumping and landing in slow motion against a black background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 33.0], "base_se_norm": [-0.8941163420677185, -0.563230037689209], "gt_se_raw": [25.0, 32.0], "gt_se_norm": [-0.6691136360168457, -0.5764654874801636], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003039.pt"}
{"idx": 3041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MxNfvh7vaSs.mp4", "duration": 149.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the user selecting the game level?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the user selecting the game level?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 39.0], "base_se_norm": [-0.8662505745887756, -0.47837722301483154], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.6255015730857849, -0.5586268901824951], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003041.pt"}
{"idx": 3043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MxNfvh7vaSs.mp4", "duration": 149.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the two different braille display areas are highlighted and interacted with.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the two different braille display areas are highlighted and interacted with.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 137.0], "base_se_norm": [0.0031207799911499023, 0.8323673009872437], "gt_se_raw": [76.0, 89.0], "gt_se_norm": [0.016495704650878906, 0.19036996364593506], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003043.pt"}
{"idx": 3045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MxNfvh7vaSs.mp4", "duration": 149.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the user make a mistake and receive negative feedback?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the user make a mistake and receive negative feedback?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.5051270723342896, -0.46500223875045776], "gt_se_raw": [124.0, 128.0], "gt_se_norm": [0.6584930419921875, 0.7119927406311035], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003045.pt"}
{"idx": 3047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DiZEribdeTw.mp4", "duration": 51.176125, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the red shirt approach the log and stop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the red shirt approach the log and stop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.8436771035194397, -0.6873542070388794], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.8436771035194397, -0.7264349460601807], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003047.pt"}
{"idx": 3049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DiZEribdeTw.mp4", "duration": 51.176125, "prompt_text": "Please find the visual event described by the sentence 'When does the man swing the axe and successfully split the log?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man swing the axe and successfully split the log?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.296546995639801, -0.17930477857589722], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.296546995639801, -0.2183855175971985], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003049.pt"}
{"idx": 3051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QFMMD4JwTwM.mp4", "duration": 15.482133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the referee dropping the puck to start the play?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the referee dropping the puck to start the play?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.6124565005302429], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.7416377067565918, -0.4832753539085388], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003051.pt"}
{"idx": 3053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QFMMD4JwTwM.mp4", "duration": 15.482133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the player in the orange jersey shoots the puck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the player in the orange jersey shoots the puck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.4209928512573242, 0.6793551445007324], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [0.4209928512573242, 0.5501739978790283], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003053.pt"}
{"idx": 3055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QFMMD4JwTwM.mp4", "duration": 15.482133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the goalie fall to the ice in an attempt to block the puck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the goalie fall to the ice in an attempt to block the puck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.4209928512573242, 0.6793551445007324], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [0.5501739978790283, 0.6793551445007324], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003055.pt"}
{"idx": 3057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txqiwrbYGrs.mp4", "duration": 118.9, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the boy mentions he has 'two fingers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the boy mentions he has 'two fingers'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.42809081077575684, -0.32716572284698486], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.5458368062973022, -0.5121951103210449], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003057.pt"}
{"idx": 3059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txqiwrbYGrs.mp4", "duration": 118.9, "prompt_text": "Please find the visual event described by the sentence 'When does the boy say he can't see anything?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy say he can't see anything?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 53.0], "base_se_norm": [-0.20941966772079468, -0.10849452018737793], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.19259881973266602, -0.14213627576828003], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003059.pt"}
{"idx": 3061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/twKxRnoGxoM.mp4", "duration": 65.73233333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a graduation photo of a young woman and an old man is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a graduation photo of a young woman and an old man is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.9391471147537231, -0.817441463470459], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.9087207317352295, -0.817441463470459], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003061.pt"}
{"idx": 3063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/twKxRnoGxoM.mp4", "duration": 65.73233333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the Facebook comments being displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the Facebook comments being displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.17848646640777588, 0.03449857234954834], "gt_se_raw": [27.0, 35.0], "gt_se_norm": [-0.17848646640777588, 0.06492495536804199], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003063.pt"}
{"idx": 3065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/twKxRnoGxoM.mp4", "duration": 65.73233333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man wearing a blue vest with a red 'B' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man wearing a blue vest with a red 'B' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.4604685306549072, 0.5821743011474609], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.4604685306549072, 0.5821743011474609], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003065.pt"}
{"idx": 3067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vy6qB2FLbvE.mp4", "duration": 104.31253333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the meatballs being placed on the baking rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the meatballs being placed on the baking rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.559017539024353, -0.5206711888313293], "gt_se_raw": [24.0, 25.0], "gt_se_norm": [-0.5398443937301636, -0.5206711888313293], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003067.pt"}
{"idx": 3069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vy6qB2FLbvE.mp4", "duration": 104.31253333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the stuffed shells being arranged in the baking dish with tomato sauce?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the stuffed shells being arranged in the baking dish with tomato sauce?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [0.016177058219909668, 0.11204278469085693], "gt_se_raw": [53.0, 58.0], "gt_se_norm": [0.016177058219909668, 0.11204278469085693], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003069.pt"}
{"idx": 3071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lyoaLway15g.mp4", "duration": 199.29166666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the girl looking at her reflection in the glass of the record store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the girl looking at her reflection in the glass of the record store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 24.0], "base_se_norm": [-0.9297512173652649, -0.7591469883918762], "gt_se_raw": [6.0, 10.0], "gt_se_norm": [-0.9397867321968079, -0.8996445536613464], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003071.pt"}
{"idx": 3073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lyoaLway15g.mp4", "duration": 199.29166666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the girl puts on headphones and starts listening to music while lying on her bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the girl puts on headphones and starts listening to music while lying on her bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 71.0], "base_se_norm": [-0.37779635190963745, -0.28747648000717163], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.3376542329788208, -0.28747648000717163], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003073.pt"}
{"idx": 3075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lyoaLway15g.mp4", "duration": 199.29166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the girl, looking sad, takes down the poster of her idol from the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the girl, looking sad, takes down the poster of her idol from the mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 138.0], "base_se_norm": [0.31465601921081543, 0.3849048614501953], "gt_se_raw": [137.0, 141.0], "gt_se_norm": [0.37486934661865234, 0.4150114059448242], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003075.pt"}
{"idx": 3077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WO73wnquXtM.mp4", "duration": 187.7709111111111, "prompt_text": "Please find the visual event described by the sentence 'At what point are fruits and flowers being added to the blue smoothie bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are fruits and flowers being added to the blue smoothie bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 55.0], "base_se_norm": [-0.4993899464607239, -0.4141797423362732], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.45678484439849854, -0.4248310327529907], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003077.pt"}
{"idx": 3079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WO73wnquXtM.mp4", "duration": 187.7709111111111, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the pink smoothie is poured into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the pink smoothie is poured into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 142.0], "base_se_norm": [0.459225058555603, 0.5124814510345459], "gt_se_raw": [138.0, 142.0], "gt_se_norm": [0.4698762893676758, 0.5124814510345459], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003079.pt"}
{"idx": 3081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S7VYhBwbprE.mp4", "duration": 97.2972, "prompt_text": "Please find the visual event described by the sentence 'When does the man first use the trackpad on the laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man first use the trackpad on the laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 17.0], "base_se_norm": [-0.9794444441795349, -0.6505552530288696], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9794444441795349, -0.9177777171134949], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003081.pt"}
{"idx": 3083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S7VYhBwbprE.mp4", "duration": 97.2972, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the man holds his right hand up to emphasize a point about percentages.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the man holds his right hand up to emphasize a point about percentages.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [-0.32166600227355957, -0.1777769923210144], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.21888810396194458, -0.13666582107543945], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003083.pt"}
{"idx": 3085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S7VYhBwbprE.mp4", "duration": 97.2972, "prompt_text": "Please find the visual event described by the sentence 'Locate the exchange where the woman explains she cannot do the man's work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the exchange where the woman explains she cannot do the man's work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 78.0], "base_se_norm": [0.27444565296173096, 0.603334903717041], "gt_se_raw": [87.0, 96.0], "gt_se_norm": [0.7883350849151611, 0.9733352661132812], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003085.pt"}
{"idx": 3087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nTgVwn0ZrXM.mp4", "duration": 39.039, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman is rolling the baby's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman is rolling the baby's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.3852301836013794, -0.07784527540206909], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.2827685475349426, -0.07784527540206909], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003087.pt"}
{"idx": 3089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nTgVwn0ZrXM.mp4", "duration": 39.039, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the woman guides the baby's hand to mark a 'W' in the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the woman guides the baby's hand to mark a 'W' in the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.3339993357658386, -0.07784527540206909], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.2295396327972412, 0.33200132846832275], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003089.pt"}
{"idx": 3091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtS1l2NXWX8.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'When can we see the phrase 'Help notice danger' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the phrase 'Help notice danger' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.836341142654419, -0.8090646266937256], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.836341142654419, -0.8090646266937256], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003091.pt"}
{"idx": 3093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtS1l2NXWX8.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'At what point is a guide dog shown assisting a person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a guide dog shown assisting a person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [-0.30444979667663574, -0.27717334032058716], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.30444979667663574, -0.27717334032058716], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003093.pt"}
{"idx": 3095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtS1l2NXWX8.mp4", "duration": 146.6465, "prompt_text": "Please find the visual event described by the sentence 'When does the video show examples of tactile paving in Japan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show examples of tactile paving in Japan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 128.0], "base_se_norm": [0.4865680932998657, 0.745694637298584], "gt_se_raw": [110.0, 128.0], "gt_se_norm": [0.5002062320709229, 0.745694637298584], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003095.pt"}
{"idx": 3097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nX4mmS4k5Co.mp4", "duration": 39.666666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the cashier scans an item.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the cashier scans an item.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.04201686382293701, 0.058823466300964355], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.008403301239013672, 0.10924363136291504], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003097.pt"}
{"idx": 3099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nX4mmS4k5Co.mp4", "duration": 39.666666666666664, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the hooded character looks confused after hearing the cashier's contradictory statement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the hooded character looks confused after hearing the cashier's contradictory statement.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 19.0], "base_se_norm": [-0.3949580192565918, -0.04201686382293701], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.6638654470443726, 0.7647057771682739], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003099.pt"}
{"idx": 3101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L3MtFGWRXAA.mp4", "duration": 214.75620833333332, "prompt_text": "Please find the visual event described by the sentence 'When is the man bringing the puppy home for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man bringing the puppy home for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 27.0], "base_se_norm": [-0.8044294118881226, -0.7485520839691162], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.8044294118881226, -0.7578649520874023], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003101.pt"}
{"idx": 3103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L3MtFGWRXAA.mp4", "duration": 214.75620833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man and the puppy are playing the piano together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man and the puppy are playing the piano together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 82.0], "base_se_norm": [-0.35741090774536133, -0.2363433837890625], "gt_se_raw": [69.0, 82.0], "gt_se_norm": [-0.35741090774536133, -0.2363433837890625], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003103.pt"}
{"idx": 3105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/L3MtFGWRXAA.mp4", "duration": 214.75620833333332, "prompt_text": "Please find the visual event described by the sentence 'When does the puppy destroy the man's miniature model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 170.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppy destroy the man's miniature model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 170.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 170.0], "base_se_norm": [0.527313232421875, 0.5831905603408813], "gt_se_raw": [163.0, 167.0], "gt_se_norm": [0.5180003643035889, 0.5552518367767334], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003105.pt"}
{"idx": 3107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gLax3zOBN40.mp4", "duration": 108.9088, "prompt_text": "Please find the visual event described by the sentence 'Find the time when a man in a green apron is holding a baby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when a man in a green apron is holding a baby.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.2654404044151306, -0.2103484869003296], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.2654404044151306, -0.2103484869003296], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003107.pt"}
{"idx": 3109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gLax3zOBN40.mp4", "duration": 108.9088, "prompt_text": "Please find the visual event described by the sentence 'Show me when a presenter with curly hair is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a presenter with curly hair is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 82.0], "base_se_norm": [-0.06343656778335571, 0.5058470964431763], "gt_se_raw": [51.0, 83.0], "gt_se_norm": [-0.06343656778335571, 0.524211049079895], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003109.pt"}
{"idx": 3111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HFa9WfIJgtY.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where water is being poured for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where water is being poured for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.6753246784210205, -0.550449550151825], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.6753246784210205, -0.5754245519638062], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003111.pt"}
{"idx": 3113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HFa9WfIJgtY.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the Aeropress is flipped and placed on the cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the Aeropress is flipped and placed on the cup.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 53.0], "base_se_norm": [0.12387609481811523, 0.32367634773254395], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [0.12387609481811523, 0.2237762212753296], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003113.pt"}
{"idx": 3115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IkpWQHLY4A4.mp4", "duration": 184.5, "prompt_text": "Please find the visual event described by the sentence 'When does the student knock on the professor's door and then the professor says 'Come on in'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the student knock on the professor's door and then the professor says 'Come on in'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 26.0], "base_se_norm": [-0.783197820186615, -0.718157172203064], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.4037940502166748, -0.3712736964225769], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003115.pt"}
{"idx": 3117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IkpWQHLY4A4.mp4", "duration": 184.5, "prompt_text": "Please find the visual event described by the sentence 'At what point are two male students sitting at a table and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two male students sitting at a table and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8915989398956299, -0.859078586101532], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.8915989398956299, -0.859078586101532], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003117.pt"}
{"idx": 3119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IkpWQHLY4A4.mp4", "duration": 184.5, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman is sitting in the grass and writing in a book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman is sitting in the grass and writing in a book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.859078586101532, -0.8048780560493469], "gt_se_raw": [13.0, 19.0], "gt_se_norm": [-0.859078586101532, -0.794037938117981], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003119.pt"}
{"idx": 3121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzJqeyxye_E.mp4", "duration": 91.42466666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the moon begin to cover the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the moon begin to cover the sun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 58.0], "base_se_norm": [-0.05933481454849243, 0.2688041925430298], "gt_se_raw": [44.0, 48.0], "gt_se_norm": [-0.03745889663696289, 0.050044894218444824], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003121.pt"}
{"idx": 3123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzJqeyxye_E.mp4", "duration": 91.42466666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the period when the sun is visible only as a thin crescent shape before disappearing completely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the period when the sun is visible only as a thin crescent shape before disappearing completely.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 60.0], "base_se_norm": [0.18130040168762207, 0.31255602836608887], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [0.2031763792037964, 0.2688041925430298], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003123.pt"}
{"idx": 3125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F1NR-_YqOgE.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is significantly greater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker state that the employment rate for non-disabled Americans is significantly greater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9031662940979004, -0.4189978241920471], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9031662940979004, -0.6126651763916016], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003125.pt"}
{"idx": 3127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F1NR-_YqOgE.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is it mentioned that disabled individuals are employed at almost half the frequency of non-disabled individuals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is it mentioned that disabled individuals are employed at almost half the frequency of non-disabled individuals?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9031662940979004, -0.03166300058364868], "gt_se_raw": [10.0, 14.0], "gt_se_norm": [-0.03166300058364868, 0.355671763420105], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003127.pt"}
{"idx": 3129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F1NR-_YqOgE.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the audio segment corresponding to the phrase 'than those with disabilities'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the audio segment corresponding to the phrase 'than those with disabilities'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.03166300058364868, 0.355671763420105], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.6126651763916016, -0.4189978241920471], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003129.pt"}
{"idx": 3131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I82ffq4niLU.mp4", "duration": 75.54213333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman interact with a child by mimicking their vocalizations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman interact with a child by mimicking their vocalizations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 75.0], "base_se_norm": [0.7473691701889038, 0.9856467247009277], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [0.29728925228118896, 0.4296656847000122], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003131.pt"}
{"idx": 3133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I82ffq4niLU.mp4", "duration": 75.54213333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a man and a child are interacting at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a man and a child are interacting at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [0.6414680480957031, 0.7473691701889038], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [0.6414680480957031, 0.7473691701889038], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003133.pt"}
{"idx": 3135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oUCSXtTHo5s.mp4", "duration": 167.83433333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the paper folded in half to form a triangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the paper folded in half to form a triangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 35.0], "base_se_norm": [-0.8331688046455383, -0.5829220414161682], "gt_se_raw": [14.0, 20.0], "gt_se_norm": [-0.8331688046455383, -0.7616697549819946], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003135.pt"}
{"idx": 3137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oUCSXtTHo5s.mp4", "duration": 167.83433333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the bottom corner is folded upwards to form the snout.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the bottom corner is folded upwards to form the snout.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [0.19165122509002686, 0.3346494436264038], "gt_se_raw": [100.0, 105.0], "gt_se_norm": [0.19165122509002686, 0.2512338161468506], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003137.pt"}
{"idx": 3139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oUCSXtTHo5s.mp4", "duration": 167.83433333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me when the completed origami dog face is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 148.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the completed origami dog face is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 148.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [148.0, 159.0], "base_se_norm": [0.7636438608169556, 0.8947255611419678], "gt_se_raw": [148.0, 159.0], "gt_se_norm": [0.7636438608169556, 0.8947255611419678], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003139.pt"}
{"idx": 3141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hoQ0_smwA3k.mp4", "duration": 69.53613333333334, "prompt_text": "Please find the visual event described by the sentence 'Where can I find the tip 'Do Other Things with Your Hands'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find the tip 'Do Other Things with Your Hands'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 38.0], "base_se_norm": [-0.19466328620910645, 0.09295690059661865], "gt_se_raw": [27.0, 37.0], "gt_se_norm": [-0.22342532873153687, 0.06419491767883301], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003141.pt"}
{"idx": 3143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hoQ0_smwA3k.mp4", "duration": 69.53613333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a lotus flower icon is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a lotus flower icon is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 60.0], "base_se_norm": [0.3805772066116333, 0.7257214784622192], "gt_se_raw": [47.0, 60.0], "gt_se_norm": [0.35181522369384766, 0.7257214784622192], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003143.pt"}
{"idx": 3145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F7Ug863S8dQ.mp4", "duration": 163.20470833333331, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in a black dress stand with a red balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in a black dress stand with a red balloon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.5465817451477051, -0.4730544090270996], "gt_se_raw": [36.0, 43.0], "gt_se_norm": [-0.5588362812995911, -0.4730544090270996], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003145.pt"}
{"idx": 3147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F7Ug863S8dQ.mp4", "duration": 163.20470833333331, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man in a tuxedo is engulfed in a large cloud of smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man in a tuxedo is engulfed in a large cloud of smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.6323635578155518, -0.6078544855117798], "gt_se_raw": [31.0, 32.0], "gt_se_norm": [-0.6201090216636658, -0.6078544855117798], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003147.pt"}
{"idx": 3149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F7Ug863S8dQ.mp4", "duration": 163.20470833333331, "prompt_text": "Please find the visual event described by the sentence 'When can we see the frog being brought back to life with electricity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the frog being brought back to life with electricity?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 120.0], "base_se_norm": [0.3847639560699463, 0.47054576873779297], "gt_se_raw": [116.0, 121.0], "gt_se_norm": [0.421527624130249, 0.48280036449432373], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003149.pt"}
{"idx": 3151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y_w105aWPNY.mp4", "duration": 157.94258373205741, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man covered in snow enter the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man covered in snow enter the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 51.0], "base_se_norm": [-0.41750985383987427, -0.3541957139968872], "gt_se_raw": [46.0, 49.0], "gt_se_norm": [-0.41750985383987427, -0.37952136993408203], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003151.pt"}
{"idx": 3153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y_w105aWPNY.mp4", "duration": 157.94258373205741, "prompt_text": "Please find the visual event described by the sentence 'When does the shopkeeper show a picture of his family in the sauna?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the shopkeeper show a picture of his family in the sauna?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 106.0], "base_se_norm": [0.2916085720062256, 0.34225988388061523], "gt_se_raw": [102.0, 104.0], "gt_se_norm": [0.2916085720062256, 0.3169342279434204], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003153.pt"}
{"idx": 3155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xrhkfADEtMU.mp4", "duration": 147.52237499999998, "prompt_text": "Please find the visual event described by the sentence 'When does SpongeBob demonstrate opening the jar for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does SpongeBob demonstrate opening the jar for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9864427447319031, -0.9457709193229675], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9864427447319031, -0.9457709193229675], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003155.pt"}
{"idx": 3157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xrhkfADEtMU.mp4", "duration": 147.52237499999998, "prompt_text": "Please find the visual event described by the sentence 'Show me when Patrick is panicking and thinks he is on fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when Patrick is panicking and thinks he is on fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 49.0], "base_se_norm": [-0.5526102185249329, -0.33569395542144775], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.3628085255622864, -0.32213670015335083], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003157.pt"}
{"idx": 3159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xrhkfADEtMU.mp4", "duration": 147.52237499999998, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when Patrick emerges from his house dressed exactly like SpongeBob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when Patrick emerges from his house dressed exactly like SpongeBob.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 123.0], "base_se_norm": [0.6268719434738159, 0.6675437688827515], "gt_se_raw": [121.0, 124.0], "gt_se_norm": [0.6404291391372681, 0.6811009645462036], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003159.pt"}
{"idx": 3161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6n5IGnuadhM.mp4", "duration": 92.994, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a butterfly lands on the woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a butterfly lands on the woman's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.3547971248626709, -0.29027682542800903], "gt_se_raw": [29.0, 34.0], "gt_se_norm": [-0.37630385160446167, -0.2687700390815735], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003161.pt"}
{"idx": 3163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6n5IGnuadhM.mp4", "duration": 92.994, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the blue man hugs the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the blue man hugs the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 79.0], "base_se_norm": [0.41944634914398193, 0.6990343332290649], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [0.6345139741897583, 0.7420477867126465], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003163.pt"}
{"idx": 3165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gyktHss8qmU.mp4", "duration": 59.893166666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the robot move under the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the robot move under the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 11.0], "base_se_norm": [-0.766250491142273, -0.632679283618927], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.766250491142273, -0.632679283618927], "start_token_span": [63, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003165.pt"}
{"idx": 3167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gyktHss8qmU.mp4", "duration": 59.893166666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part of the video that shows the cartoon inspiration for the robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part of the video that shows the cartoon inspiration for the robot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 55.0], "base_se_norm": [0.2355332374572754, 0.8366035223007202], "gt_se_raw": [36.0, 41.0], "gt_se_norm": [0.2021404504776001, 0.3691045045852661], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003167.pt"}
{"idx": 3169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gyktHss8qmU.mp4", "duration": 59.893166666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the robot maid named Rosie is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the robot maid named Rosie is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 55.0], "base_se_norm": [0.3691045045852661, 0.8366035223007202], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [0.703032374382019, 0.8032107353210449], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003169.pt"}
{"idx": 3171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o2AsIXSh2xo.mp4", "duration": 176.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the car's parking sensors are beeping erratically, indicating danger from all sides.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the car's parking sensors are beeping erratically, indicating danger from all sides.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 109.0], "base_se_norm": [0.09630334377288818, 0.23192846775054932], "gt_se_raw": [107.0, 109.0], "gt_se_norm": [0.20932424068450928, 0.23192846775054932], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003171.pt"}
{"idx": 3173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o2AsIXSh2xo.mp4", "duration": 176.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the pregnant woman undergo an ultrasound examination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the pregnant woman undergo an ultrasound examination?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 41.0], "base_se_norm": [-0.6948434114456177, -0.5366140604019165], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.6722391843795776, -0.6496350169181824], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003173.pt"}
{"idx": 3175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xyw5fd5H2j0.mp4", "duration": 58.625233333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person dabbing a white cloth on another person's pants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person dabbing a white cloth on another person's pants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, -0.4882749319076538], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.8294249773025513], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003175.pt"}
{"idx": 3177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xyw5fd5H2j0.mp4", "duration": 58.625233333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman in the black t-shirt holds and inspects the other person's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman in the black t-shirt holds and inspects the other person's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 15.0], "base_se_norm": [-0.6929649710655212, -0.4882749319076538], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.7270799875259399, -0.5223899483680725], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003177.pt"}
{"idx": 3179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xyw5fd5H2j0.mp4", "duration": 58.625233333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man in a hoodie joins the group and places his hands on the person's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man in a hoodie joins the group and places his hands on the person's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [0.7398651838302612, 0.91044020652771], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [0.8422101736068726, 0.9786702394485474], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003179.pt"}
{"idx": 3181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_10tsL37VS4.mp4", "duration": 86.58333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the child holding the orange spiky toy near their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the child holding the orange spiky toy near their face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 25.0], "base_se_norm": [-0.745909571647644, -0.4225216507911682], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.5611164569854736, -0.4687199592590332], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003181.pt"}
{"idx": 3183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_10tsL37VS4.mp4", "duration": 86.58333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the child's foot get entangled with the rainbow slinky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child's foot get entangled with the rainbow slinky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 40.0], "base_se_norm": [-0.21462947130203247, -0.07603466510772705], "gt_se_raw": [35.0, 46.0], "gt_se_norm": [-0.19153034687042236, 0.0625600814819336], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003183.pt"}
{"idx": 3185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xct-dgUvlp8.mp4", "duration": 27.133333333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the dog sleeping on its back in the banana-shaped bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the dog sleeping on its back in the banana-shaped bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 26.0], "base_se_norm": [-1.0, 0.9164619445800781], "gt_se_raw": [0.0, 27.0], "gt_se_norm": [-1.0, 0.9901720285415649], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003185.pt"}
{"idx": 3187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xct-dgUvlp8.mp4", "duration": 27.133333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the dog turns its head from side to side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the dog turns its head from side to side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [0.1056511402130127, 0.4742014408111572], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.3366093039512634, -0.11547911167144775], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003187.pt"}
{"idx": 3189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xct-dgUvlp8.mp4", "duration": 27.133333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the dog is lying still with its eyes slightly open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the dog is lying still with its eyes slightly open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.1056511402130127], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [0.25307130813598633, 0.4742014408111572], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003189.pt"}
{"idx": 3191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3YE4sWGC2AM.mp4", "duration": 33.833800000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy stands up from his chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy stands up from his chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.2906501889228821, -0.11331278085708618], "gt_se_raw": [12.0, 15.0], "gt_se_norm": [-0.2906501889228821, -0.11331278085708618], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003191.pt"}
{"idx": 3193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3YE4sWGC2AM.mp4", "duration": 33.833800000000004, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the boy is held up high, touching the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the boy is held up high, touching the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 29.0], "base_se_norm": [0.2413620948791504, 0.7142620086669922], "gt_se_raw": [22.0, 29.0], "gt_se_norm": [0.3004746437072754, 0.7142620086669922], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003193.pt"}
{"idx": 3195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tUyeaT2ZX1I.mp4", "duration": 169.23573333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a woman is moving boxes with a pallet jack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a woman is moving boxes with a pallet jack.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.7400076389312744, -0.7045540809631348], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.7400076389312744, -0.7045540809631348], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003195.pt"}
{"idx": 3197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tUyeaT2ZX1I.mp4", "duration": 169.23573333333334, "prompt_text": "Please find the visual event described by the sentence 'When is a woman in a red sweater using a power drill to assemble a tool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman in a red sweater using a power drill to assemble a tool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 57.0], "base_se_norm": [-0.36183691024780273, -0.32638341188430786], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.36183691024780273, -0.32638341188430786], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003197.pt"}
{"idx": 3199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tUyeaT2ZX1I.mp4", "duration": 169.23573333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a part is being inserted into a heating machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a part is being inserted into a heating machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 106.0], "base_se_norm": [0.21723699569702148, 0.25269055366516113], "gt_se_raw": [103.0, 105.0], "gt_se_norm": [0.21723699569702148, 0.24087274074554443], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003199.pt"}
{"idx": 3201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ysm7LT3ojDQ.mp4", "duration": 156.23941666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the father walks upstairs to check on the noise for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the father walks upstairs to check on the noise for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [-0.5519696474075317, -0.5007661581039429], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.5519696474075317, -0.5263679027557373], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003201.pt"}
{"idx": 3203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ysm7LT3ojDQ.mp4", "duration": 156.23941666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the mother run upstairs to grab her daughter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the mother run upstairs to grab her daughter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 135.0], "base_se_norm": [0.6897145509719849, 0.7281171083450317], "gt_se_raw": [133.0, 134.0], "gt_se_norm": [0.7025153636932373, 0.7153162956237793], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003203.pt"}
{"idx": 3205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zC2G6lf9fCs.mp4", "duration": 60.06, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man in the suit is led into the examination room by the nurse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man in the suit is led into the examination room by the nurse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8667998909950256], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9666999578475952, -0.9000998735427856], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003205.pt"}
{"idx": 3207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zC2G6lf9fCs.mp4", "duration": 60.06, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man wipes down the intercom phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man wipes down the intercom phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.20079922676086426, -0.06759911775588989], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.16749918460845947, -0.1008991003036499], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003207.pt"}
{"idx": 3209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zC2G6lf9fCs.mp4", "duration": 60.06, "prompt_text": "Please find the visual event described by the sentence 'When does the doctor inform the elderly couple of the surprising news?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the doctor inform the elderly couple of the surprising news?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 59.0], "base_se_norm": [0.898101806640625, 0.9647018909454346], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [0.7982017993927002, 0.9314018487930298], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003209.pt"}
{"idx": 3211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7wJRtJvcyto.mp4", "duration": 150.31683333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man with striking blue eyes looking up in shock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man with striking blue eyes looking up in shock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.8004215359687805, -0.7605058550834656], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.8004215359687805, -0.7605058550834656], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003211.pt"}
{"idx": 3213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7wJRtJvcyto.mp4", "duration": 150.31683333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a man in armor is screaming in the rain next to a wooden cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a man in armor is screaming in the rain next to a wooden cart.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [-0.5609273910522461, -0.5210117101669312], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.5609273910522461, -0.5210117101669312], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003213.pt"}
{"idx": 3215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZGA5yi-_QHM.mp4", "duration": 81.68, "prompt_text": "Please find the visual event described by the sentence 'When does the female toy slide down the tent and swing on the lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the female toy slide down the tent and swing on the lights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.5592555999755859, -0.3633692264556885], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [-0.4857982397079468, -0.38785505294799805], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003215.pt"}
{"idx": 3217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZGA5yi-_QHM.mp4", "duration": 81.68, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the female toy walks on the beam of a Ferris wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the female toy walks on the beam of a Ferris wheel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 41.0], "base_se_norm": [-0.3143976330757141, 0.003917694091796875], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [-0.14299708604812622, -0.02056807279586792], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003217.pt"}
{"idx": 3219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZGA5yi-_QHM.mp4", "duration": 81.68, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where the red balloon floats up towards the moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where the red balloon floats up towards the moon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [0.44466209411621094, 0.5915769338607788], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [0.4691479206085205, 0.5915769338607788], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003219.pt"}
{"idx": 3221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7IhJBWv4Qx8.mp4", "duration": 154.88806666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a boy in glasses looking intently at a lit-up screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a boy in glasses looking intently at a lit-up screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 104.0], "base_se_norm": [0.2912551164627075, 0.34290528297424316], "gt_se_raw": [100.0, 105.0], "gt_se_norm": [0.2912551164627075, 0.35581791400909424], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003221.pt"}
{"idx": 3223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7IhJBWv4Qx8.mp4", "duration": 154.88806666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a boy in a wheelchair is shown with a device on his lap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a boy in a wheelchair is shown with a device on his lap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 79.0], "base_se_norm": [-0.05738377571105957, 0.020091533660888672], "gt_se_raw": [72.0, 79.0], "gt_se_norm": [-0.07029634714126587, 0.020091533660888672], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003223.pt"}
{"idx": 3225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PmFyn_0udo8.mp4", "duration": 469.84437499999996, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a shopping cart is shown on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a shopping cart is shown on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [-0.7105424404144287, -0.6977722644805908], "gt_se_raw": [68.0, 72.0], "gt_se_norm": [-0.7105424404144287, -0.6935155391693115], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003225.pt"}
{"idx": 3227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PmFyn_0udo8.mp4", "duration": 469.84437499999996, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the word 'Conclusion' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the word 'Conclusion' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 58.0], "base_se_norm": [-0.7829068303108215, -0.7531097531318665], "gt_se_raw": [51.0, 57.0], "gt_se_norm": [-0.7829068303108215, -0.7573664784431458], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003227.pt"}
{"idx": 3229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PmFyn_0udo8.mp4", "duration": 469.84437499999996, "prompt_text": "Please find the visual event described by the sentence 'Identify the segment where an accessibility meter is animated on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the segment where an accessibility meter is animated on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 168.0], "base_se_norm": [-0.3146666884422302, -0.28486955165863037], "gt_se_raw": [161.0, 168.0], "gt_se_norm": [-0.3146666884422302, -0.28486955165863037], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003229.pt"}
{"idx": 3231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8pfPl8BkfVY.mp4", "duration": 122.95616666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman hug the man from behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman hug the man from behind?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.31682974100112915, -0.2517659068107605], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [-0.31682974100112915, -0.2517659068107605], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003231.pt"}
{"idx": 3233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8pfPl8BkfVY.mp4", "duration": 122.95616666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman gives the man a kiss on the cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman gives the man a kiss on the cheek.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [-0.31682974100112915, -0.2517659068107605], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [0.05728733539581299, 0.08981919288635254], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003233.pt"}
{"idx": 3235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cYNdUM2gRsg.mp4", "duration": 93.76, "prompt_text": "Please find the visual event described by the sentence 'When does the character say 'I wanna be a pie' and get into the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character say 'I wanna be a pie' and get into the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.3387371897697449, -0.27474403381347656], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [-0.3387371897697449, -0.29607510566711426], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003235.pt"}
{"idx": 3237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cYNdUM2gRsg.mp4", "duration": 93.76, "prompt_text": "Please find the visual event described by the sentence 'At what point does one character steal another character's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one character steal another character's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 64.0], "base_se_norm": [0.30119454860687256, 0.3651876449584961], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [-0.18941980600357056, -0.14675772190093994], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003237.pt"}
{"idx": 3239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cYNdUM2gRsg.mp4", "duration": 93.76, "prompt_text": "Please find the visual event described by the sentence 'When does the character get kicked into the sky after saying she wants to go to the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character get kicked into the sky after saying she wants to go to the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9786689281463623, -0.9360068440437317], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.8131399154663086, 0.855802059173584], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003239.pt"}
{"idx": 3241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ALXYLq5STlM.mp4", "duration": 43.076366666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman raises the child's hands high up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman raises the child's hands high up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [0.30001676082611084, 0.4857335090637207], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [0.30001676082611084, 0.4857335090637207], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003241.pt"}
{"idx": 3243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ALXYLq5STlM.mp4", "duration": 43.076366666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the child wearing a black and white striped shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the child wearing a black and white striped shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 42.0], "base_se_norm": [-1.0, 0.950025200843811], "gt_se_raw": [0.0, 43.0], "gt_se_norm": [-1.0, 0.9964543581008911], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003243.pt"}
{"idx": 3245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fuj29nJezzc.mp4", "duration": 335.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the player with the number 73 on their back while the scoreboard shows NYG 26 and DAL 20?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the player with the number 73 on their back while the scoreboard shows NYG 26 and DAL 20?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [-0.7255097031593323, -0.689706563949585], "gt_se_raw": [51.0, 53.0], "gt_se_norm": [-0.6956738233566284, -0.683739423751831], "start_token_span": [84, 88], "end_token_span": [90, 94], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003245.pt"}
{"idx": 3247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fuj29nJezzc.mp4", "duration": 335.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'A player in a white jersey is tackled by two players in blue jerseys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A player in a white jersey is tackled by two players in blue jerseys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 65.0], "base_se_norm": [-0.6419691443443298, -0.612133264541626], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.6360019445419312, -0.612133264541626], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003247.pt"}
{"idx": 3249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Fuj29nJezzc.mp4", "duration": 335.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the players in white jerseys start to celebrate after the touchdown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 198.0 - 204.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the players in white jerseys start to celebrate after the touchdown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 198.0 - 204.0 seconds.<|im_end|>\n", "base_se_raw": [198.0, 204.0], "base_se_norm": [0.18150174617767334, 0.21730482578277588], "gt_se_raw": [199.0, 204.0], "gt_se_norm": [0.1874690055847168, 0.21730482578277588], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003249.pt"}
{"idx": 3251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbtsKmUMVhU.mp4", "duration": 19.586233333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person successfully lift the white disc off the floor with the tool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person successfully lift the white disc off the floor with the tool?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [0.1232379674911499, 0.838025689125061], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [0.021125435829162598, 0.2253504991531372], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003251.pt"}
{"idx": 3253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fbtsKmUMVhU.mp4", "duration": 19.586233333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the person on the left leans in to get a closer look at the action.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the person on the left leans in to get a closer look at the action.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [0.3274630308151245, 0.838025689125061], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [0.5316880941390991, 0.9401382207870483], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003253.pt"}
{"idx": 3255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5xJ6h6OTveU.mp4", "duration": 82.48248248248248, "prompt_text": "Please find the visual event described by the sentence 'When does the player in the black jersey make the game-winning shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player in the black jersey make the game-winning shot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.7090291380882263, -0.6362863779067993], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.684781551361084, -0.6362863779067993], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003255.pt"}
{"idx": 3257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5xJ6h6OTveU.mp4", "duration": 82.48248248248248, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the disappointed reaction from the players in the white jerseys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the disappointed reaction from the players in the white jerseys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.5275970697402954, 0.6488349437713623], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.2725728154182434, -0.19983011484146118], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003257.pt"}
{"idx": 3259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5xJ6h6OTveU.mp4", "duration": 82.48248248248248, "prompt_text": "Please find the visual event described by the sentence 'When are the celebrating fans shown up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the celebrating fans shown up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [-0.19983011484146118, -0.07859224081039429], "gt_se_raw": [37.0, 40.0], "gt_se_norm": [-0.10283982753753662, -0.03009706735610962], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003259.pt"}
{"idx": 3261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S8hUzi6v43c.mp4", "duration": 189.23070833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when they cut the candy into about 3,000 pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when they cut the candy into about 3,000 pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 125.0], "base_se_norm": [0.2894313335418701, 0.3211386203765869], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.2894313335418701, 0.3211386203765869], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003261.pt"}
{"idx": 3263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S8hUzi6v43c.mp4", "duration": 189.23070833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me a close-up of the finished koala candy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me a close-up of the finished koala candy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 135.0], "base_se_norm": [0.37398409843444824, 0.4268296957015991], "gt_se_raw": [131.0, 135.0], "gt_se_norm": [0.38455331325531006, 0.4268296957015991], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003263.pt"}
{"idx": 3265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W9fRvESVClE.mp4", "duration": 106.29166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the zombie walks down the ruined street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the zombie walks down the ruined street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.7742061614990234, -0.6801254749298096], "gt_se_raw": [12.0, 17.0], "gt_se_norm": [-0.7742061614990234, -0.6801254749298096], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003265.pt"}
{"idx": 3267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W9fRvESVClE.mp4", "duration": 106.29166666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the dog lies on its back and wags its tail.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the dog lies on its back and wags its tail.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 78.0], "base_se_norm": [0.24186599254608154, 0.4676598310470581], "gt_se_raw": [42.0, 45.0], "gt_se_norm": [-0.20972168445587158, -0.15327322483062744], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003267.pt"}
{"idx": 3269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W9fRvESVClE.mp4", "duration": 106.29166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the zombie scratches the dog's belly, turning it into a zombie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the zombie scratches the dog's belly, turning it into a zombie.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 78.0], "base_se_norm": [0.2606821060180664, 0.4676598310470581], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [0.3735790252685547, 0.4676598310470581], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003269.pt"}
{"idx": 3271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_F6h0yH7EyU.mp4", "duration": 145.25, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the Grinch throws things while in bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the Grinch throws things while in bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.49053359031677246, -0.4079173803329468], "gt_se_raw": [34.0, 40.0], "gt_se_norm": [-0.5318416357040405, -0.4492254853248596], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003271.pt"}
{"idx": 3273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_F6h0yH7EyU.mp4", "duration": 145.25, "prompt_text": "Please find the visual event described by the sentence 'When does the dog get clothes for the Grinch from the closet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog get clothes for the Grinch from the closet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 83.0], "base_se_norm": [0.08777964115142822, 0.14285719394683838], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.08777964115142822, 0.14285719394683838], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003273.pt"}
{"idx": 3275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICVXf8Vznec.mp4", "duration": 169.40256666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two girls arguing in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two girls arguing in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 19.0], "base_se_norm": [-0.9645814299583435, -0.775682270526886], "gt_se_raw": [4.0, 19.0], "gt_se_norm": [-0.952775239944458, -0.775682270526886], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003275.pt"}
{"idx": 3277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICVXf8Vznec.mp4", "duration": 169.40256666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the girl in the red overalls walks down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the girl in the red overalls walks down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 49.0], "base_se_norm": [-0.50413978099823, -0.4214963912963867], "gt_se_raw": [42.0, 48.0], "gt_se_norm": [-0.50413978099823, -0.4333025813102722], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003277.pt"}
{"idx": 3279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ICVXf8Vznec.mp4", "duration": 169.40256666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the girl in the yellow shirt happily announces her news.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the girl in the yellow shirt happily announces her news.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 131.0], "base_se_norm": [0.40493738651275635, 0.5466116666793823], "gt_se_raw": [118.0, 129.0], "gt_se_norm": [0.3931311368942261, 0.5229992866516113], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003279.pt"}
{"idx": 3281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/baNaqhoBjUE.mp4", "duration": 198.96543333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman opens the green book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman opens the green book.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.8492200374603271, -0.788908064365387], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [-0.8391680717468262, -0.788908064365387], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003281.pt"}
{"idx": 3283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/baNaqhoBjUE.mp4", "duration": 198.96543333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the woman helps the boy identify the 'Shepherd' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the woman helps the boy identify the 'Shepherd' symbol.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 114.0], "base_se_norm": [-0.1154242753982544, 0.14592766761779785], "gt_se_raw": [99.0, 114.0], "gt_se_norm": [-0.004852294921875, 0.14592766761779785], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003283.pt"}
{"idx": 3285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vbuq7w3ZDUQ.mp4", "duration": 416.99991666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the moon appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the moon appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 100.0], "base_se_norm": [-0.5731413960456848, -0.5203835964202881], "gt_se_raw": [89.0, 99.0], "gt_se_norm": [-0.5731413960456848, -0.5251797437667847], "start_token_span": [63, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003285.pt"}
{"idx": 3287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vbuq7w3ZDUQ.mp4", "duration": 416.99991666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point in time does a very large star fall onto the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 262.0 - 268.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in time does a very large star fall onto the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 262.0 - 268.0 seconds.<|im_end|>\n", "base_se_raw": [262.0, 268.0], "base_se_norm": [0.25659501552581787, 0.2853720188140869], "gt_se_raw": [258.0, 262.0], "gt_se_norm": [0.237410306930542, 0.25659501552581787], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003287.pt"}
{"idx": 3289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vbuq7w3ZDUQ.mp4", "duration": 416.99991666666665, "prompt_text": "Please find the visual event described by the sentence 'When do all three of them work together to clean the stars from the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 343.0 - 358.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all three of them work together to clean the stars from the moon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 343.0 - 358.0 seconds.<|im_end|>\n", "base_se_raw": [343.0, 358.0], "base_se_norm": [0.6450842618942261, 0.7170267105102539], "gt_se_raw": [351.0, 357.0], "gt_se_norm": [0.6834535598754883, 0.7122305631637573], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003289.pt"}
{"idx": 3291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xl6yy6a3emw.mp4", "duration": 170.5704, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man is listening to the message while sitting on the sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man is listening to the message while sitting on the sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 77.0], "base_se_norm": [-0.8592956066131592, -0.09714698791503906], "gt_se_raw": [13.0, 77.0], "gt_se_norm": [-0.8475703001022339, -0.09714698791503906], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003291.pt"}
{"idx": 3293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Xl6yy6a3emw.mp4", "duration": 170.5704, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman show up at the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman show up at the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [0.18426167964935303, 0.2311631441116333], "gt_se_raw": [101.0, 103.0], "gt_se_norm": [0.18426167964935303, 0.20771241188049316], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003293.pt"}
{"idx": 3295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p_Rrovk5nsk.mp4", "duration": 64.6, "prompt_text": "Please find the visual event described by the sentence 'When is the person taking the cup from the dish rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person taking the cup from the dish rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.8761609792709351, -0.7832817435264587], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.8452012538909912, -0.7832817435264587], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003295.pt"}
{"idx": 3297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p_Rrovk5nsk.mp4", "duration": 64.6, "prompt_text": "Please find the visual event described by the sentence 'During which period is the person watering the plants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which period is the person watering the plants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 51.0], "base_se_norm": [-0.13312691450119019, 0.5789474248886108], "gt_se_raw": [29.0, 51.0], "gt_se_norm": [-0.10216718912124634, 0.5789474248886108], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003297.pt"}
{"idx": 3299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p_Rrovk5nsk.mp4", "duration": 64.6, "prompt_text": "Please find the visual event described by the sentence 'When is the cup being placed back into the dish rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the cup being placed back into the dish rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [0.6718267202377319, 0.7956656217575073], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [0.7337461709976196, 0.7956656217575073], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003299.pt"}
{"idx": 3301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzZuNeRdFIc.mp4", "duration": 134.759625, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man is opening the milk jug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man is opening the milk jug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 45.0], "base_se_norm": [-0.5250803232192993, -0.33214420080184937], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.4953978657722473, -0.46571534872055054], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003301.pt"}
{"idx": 3303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TzZuNeRdFIc.mp4", "duration": 134.759625, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range where the man is chugging the milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range where the man is chugging the milk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 105.0], "base_se_norm": [-0.1540493369102478, 0.5583301782608032], "gt_se_raw": [58.0, 104.0], "gt_se_norm": [-0.13920807838439941, 0.5434889793395996], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003303.pt"}
{"idx": 3305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vsmBpaVqhmc.mp4", "duration": 181.01416666666668, "prompt_text": "Please find the visual event described by the sentence 'Show a person holding up a book titled 'Planet Earth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show a person holding up a book titled 'Planet Earth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9668534398078918, -0.9226579666137695], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9558045864105225, -0.9337068200111389], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003305.pt"}
{"idx": 3307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vsmBpaVqhmc.mp4", "duration": 181.01416666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman is presenting the cutaway model of the space shuttle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman is presenting the cutaway model of the space shuttle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 95.0], "base_se_norm": [-0.46965473890304565, 0.04964160919189453], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.45860588550567627, -0.4254593253135681], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003307.pt"}
{"idx": 3309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vsmBpaVqhmc.mp4", "duration": 181.01416666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man attaches the inflated balloon to the string.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man attaches the inflated balloon to the string.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 153.0], "base_se_norm": [0.6241823434829712, 0.6904754638671875], "gt_se_raw": [147.0, 153.0], "gt_se_norm": [0.6241823434829712, 0.6904754638671875], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003309.pt"}
{"idx": 3311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qCNodlSc6Hw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman is using a micropipette.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman is using a micropipette.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.4753605127334595, -0.14746087789535522], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003311.pt"}
{"idx": 3313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qCNodlSc6Hw.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a woman in a lab coat smiles at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a woman in a lab coat smiles at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003313.pt"}
{"idx": 3315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9qlF_9PNhJs.mp4", "duration": 138.04, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the man climbing the ladder to save the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the man climbing the ladder to save the cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 36.0], "base_se_norm": [-0.7536945343017578, -0.47841203212738037], "gt_se_raw": [16.0, 32.0], "gt_se_norm": [-0.7681831121444702, -0.5363662242889404], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003315.pt"}
{"idx": 3317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9qlF_9PNhJs.mp4", "duration": 138.04, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cat walks on the man's back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cat walks on the man's back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 65.0], "base_se_norm": [-0.24659514427185059, -0.05824393033981323], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.21761804819107056, -0.17415237426757812], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003317.pt"}
{"idx": 3319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9qlF_9PNhJs.mp4", "duration": 138.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the cat scratches the tree trunk while the man is clinging to it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the cat scratches the tree trunk while the man is clinging to it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 45.0], "base_se_norm": [-0.5363662242889404, -0.3480150103569031], "gt_se_raw": [105.0, 108.0], "gt_se_norm": [0.5212982892990112, 0.5647639036178589], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003319.pt"}
{"idx": 3321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLdbbZZRvtM.mp4", "duration": 37.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man describe reviewing his calendar the night before?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man describe reviewing his calendar the night before?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.35368049144744873, -0.13824063539505005], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.4075404405593872, -0.245960533618927], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003321.pt"}
{"idx": 3323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLdbbZZRvtM.mp4", "duration": 37.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where he talks about arranging transportation in case he cannot drive.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where he talks about arranging transportation in case he cannot drive.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.03052067756652832, 0.23877906799316406], "gt_se_raw": [26.0, 30.0], "gt_se_norm": [0.40035903453826904, 0.615798830986023], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003323.pt"}
{"idx": 3325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JerVrbLldXw.mp4", "duration": 150.275125, "prompt_text": "Please find the visual event described by the sentence 'When is the scrawny man getting beat up in the alley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the scrawny man getting beat up in the alley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 32.0], "base_se_norm": [-0.6672769784927368, -0.5741145014762878], "gt_se_raw": [26.0, 32.0], "gt_se_norm": [-0.6539680361747742, -0.5741145014762878], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003325.pt"}
{"idx": 3327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JerVrbLldXw.mp4", "duration": 150.275125, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man start chasing the car on foot after the explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man start chasing the car on foot after the explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 108.0], "base_se_norm": [0.39743685722351074, 0.4373636245727539], "gt_se_raw": [94.0, 96.0], "gt_se_norm": [0.25103867053985596, 0.27765655517578125], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003327.pt"}
{"idx": 3329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JerVrbLldXw.mp4", "duration": 150.275125, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment Captain America uses his shield to block a flamethrower.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment Captain America uses his shield to block a flamethrower.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 127.0], "base_se_norm": [0.6636152267456055, 0.6902331113815308], "gt_se_raw": [125.0, 127.0], "gt_se_norm": [0.6636152267456055, 0.6902331113815308], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003329.pt"}
{"idx": 3331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epKCqDN9fBo.mp4", "duration": 190.39854166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the team of girls huddles around the dog in a circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the team of girls huddles around the dog in a circle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 99.0], "base_se_norm": [0.018915295600891113, 0.039923906326293945], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.018915295600891113, 0.039923906326293945], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003331.pt"}
{"idx": 3333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/epKCqDN9fBo.mp4", "duration": 190.39854166666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a close-up of the sign 'Please do not pet her!' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a close-up of the sign 'Please do not pet her!' is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 78.0], "base_se_norm": [-0.20167458057403564, -0.1806660294532776], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [-0.20167458057403564, -0.1806660294532776], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003333.pt"}
{"idx": 3335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vxd9W0RVgFY.mp4", "duration": 91.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man gets out of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man gets out of the car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.6704503893852234, -0.6045404672622681], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.6704503893852234, -0.6045404672622681], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003335.pt"}
{"idx": 3337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vxd9W0RVgFY.mp4", "duration": 91.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man interacting with a young boy wearing a red and white striped shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man interacting with a young boy wearing a red and white striped shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.1212010383605957, -0.05529111623764038], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.1212010383605957, -0.05529111623764038], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003337.pt"}
{"idx": 3339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vxd9W0RVgFY.mp4", "duration": 91.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the player runs out onto the field from the dugout.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the player runs out onto the field from the dugout.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [0.691688060760498, 0.8015378713607788], "gt_se_raw": [78.0, 82.0], "gt_se_norm": [0.7136579751968384, 0.8015378713607788], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003339.pt"}
{"idx": 3341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-wc7bB2G94.mp4", "duration": 118.15136666666668, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a blue tank top adjust the collar of the blue jacket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a blue tank top adjust the collar of the blue jacket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.5768140554428101, -0.5091042518615723], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.5768140554428101, -0.5260317325592041], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003341.pt"}
{"idx": 3343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r-wc7bB2G94.mp4", "duration": 118.15136666666668, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman in the blue jacket put her hands into her pockets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman in the blue jacket put her hands into her pockets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 60.0], "base_se_norm": [-0.03513598442077637, 0.01564633846282959], "gt_se_raw": [72.0, 76.0], "gt_se_norm": [0.21877562999725342, 0.28648531436920166], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003343.pt"}
{"idx": 3345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2imuIooWjPE.mp4", "duration": 54.72, "prompt_text": "Please find the visual event described by the sentence 'Show me when the brindle boxer dog is running.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the brindle boxer dog is running.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 12.0], "base_se_norm": [-0.8538011908531189, -0.5614035129547119], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.817251443862915, -0.6710526347160339], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003345.pt"}
{"idx": 3347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2imuIooWjPE.mp4", "duration": 54.72, "prompt_text": "Please find the visual event described by the sentence 'When can we see two dogs, one light and one dark, sniffing the ground together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two dogs, one light and one dark, sniffing the ground together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.3055555820465088, -0.23245614767074585], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.3055555820465088, -0.23245614767074585], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003347.pt"}
{"idx": 3349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2imuIooWjPE.mp4", "duration": 54.72, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a man disposes of waste in a green bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a man disposes of waste in a green bin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [0.4254385232925415, 0.5716373920440674], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [0.4985380172729492, 0.5716373920440674], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003349.pt"}
{"idx": 3351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymvdlZjewL4.mp4", "duration": 123.04166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the green shirt move a chair to sit down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the green shirt move a chair to sit down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.6261428594589233, -0.5611242651939392], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.6098882555961609, -0.5611242651939392], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003351.pt"}
{"idx": 3353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ymvdlZjewL4.mp4", "duration": 123.04166666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the doctor turns to look at the computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the doctor turns to look at the computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.5123603343963623, -0.4635963439941406], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [0.02404332160949707, 0.07280731201171875], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003353.pt"}
{"idx": 3355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q99rhZlYlAo.mp4", "duration": 110.44, "prompt_text": "Please find the visual event described by the sentence 'Show me when the cat is doing 'The Downward Cat' pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the cat is doing 'The Downward Cat' pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 20.0], "base_se_norm": [-0.7826874256134033, -0.6378123760223389], "gt_se_raw": [12.0, 19.0], "gt_se_norm": [-0.7826874256134033, -0.6559218168258667], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003355.pt"}
{"idx": 3357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q99rhZlYlAo.mp4", "duration": 110.44, "prompt_text": "Please find the visual event described by the sentence 'When can we see the 'Lazy Butterfly' pose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the 'Lazy Butterfly' pose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 43.0], "base_se_norm": [-0.3299529552459717, -0.22129666805267334], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.3299529552459717, -0.22129666805267334], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003357.pt"}
{"idx": 3359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q99rhZlYlAo.mp4", "duration": 110.44, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range for the 'Cat Tree' pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range for the 'Cat Tree' pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [0.2314378023147583, 0.32198476791381836], "gt_se_raw": [68.0, 73.0], "gt_se_norm": [0.2314378023147583, 0.32198476791381836], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003359.pt"}
{"idx": 3361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_JmA2ClUvUY.mp4", "duration": 127.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a baby with his back to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a baby with his back to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.8436685800552368], "gt_se_raw": [113.0, 116.0], "gt_se_norm": [0.7665450572967529, 0.8134444952011108], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003361.pt"}
{"idx": 3363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_JmA2ClUvUY.mp4", "duration": 127.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a baby bending over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a baby bending over?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 125.0], "base_se_norm": [0.9072433710098267, 0.9541428089141846], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.9072433710098267, 0.9541428089141846], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003363.pt"}
{"idx": 3365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mm3rTwAxH20.mp4", "duration": 70.90416666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the cartoon character type on the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cartoon character type on the computer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 14.0], "base_se_norm": [-0.8589645624160767, -0.6051007509231567], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.8307574987411499, -0.7461361885070801], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003365.pt"}
{"idx": 3367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mm3rTwAxH20.mp4", "duration": 70.90416666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket put out the fire with his foot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket put out the fire with his foot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 48.0], "base_se_norm": [0.26931893825531006, 0.35394012928009033], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [0.26931893825531006, 0.3257330656051636], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003367.pt"}
{"idx": 3369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mm3rTwAxH20.mp4", "duration": 70.90416666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman lying in bed turn on the light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman lying in bed turn on the light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [0.8334606885910034, 0.9180818796157837], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [0.8052535057067871, 0.8616677522659302], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003369.pt"}
{"idx": 3371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O8ct0a3hStg.mp4", "duration": 20.2, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person tells the cat to sit and the cat obeys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person tells the cat to sit and the cat obeys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.10891091823577881, 0.1881188154220581], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.009901046752929688, 0.1881188154220581], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003371.pt"}
{"idx": 3373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/O8ct0a3hStg.mp4", "duration": 20.2, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the person pets the black cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the person pets the black cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [0.5841583013534546, 0.8811880350112915], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [0.6831682920455933, 0.9801979064941406], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003373.pt"}
{"idx": 3375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6MW7bkk3MZ0.mp4", "duration": 141.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the red larva fall into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the red larva fall into the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.7448316812515259, -0.6881275773048401], "gt_se_raw": [66.0, 67.0], "gt_se_norm": [-0.06438273191452026, -0.05020672082901001], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003375.pt"}
{"idx": 3377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6MW7bkk3MZ0.mp4", "duration": 141.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point do both larvae become frozen solid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do both larvae become frozen solid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 88.0], "base_se_norm": [0.17660963535308838, 0.24748969078063965], "gt_se_raw": [86.0, 87.0], "gt_se_norm": [0.21913766860961914, 0.2333136796951294], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003377.pt"}
{"idx": 3379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6MW7bkk3MZ0.mp4", "duration": 141.08333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the yellow larva get hit by the red one's long tongue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the yellow larva get hit by the red one's long tongue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [0.006497383117675781, 0.0632014274597168], "gt_se_raw": [17.0, 18.0], "gt_se_norm": [-0.7590076923370361, -0.7448316812515259], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003379.pt"}
{"idx": 3381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yhofIxEfld0.mp4", "duration": 13.079733333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the liquid gets absorbed by the material.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the liquid gets absorbed by the material.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.5412750244140625, 0.0703582763671875], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.5412750244140625, -0.082550048828125], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003381.pt"}
{"idx": 3383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yhofIxEfld0.mp4", "duration": 13.079733333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the person turns the underwear inside out to show the red side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the person turns the underwear inside out to show the red side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [0.3761749267578125, 0.83489990234375], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [0.529083251953125, 0.83489990234375], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003383.pt"}
{"idx": 3385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qMzt3yQFT-Q.mp4", "duration": 397.23016666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man in the blue shirt dancing on another man's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man in the blue shirt dancing on another man's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 98.0], "base_se_norm": [-0.5468622446060181, -0.5065833330154419], "gt_se_raw": [91.0, 94.0], "gt_se_norm": [-0.5418273210525513, -0.52672278881073], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003385.pt"}
{"idx": 3387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qMzt3yQFT-Q.mp4", "duration": 397.23016666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the striped shirt take out his penis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the striped shirt take out his penis?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 76.0], "base_se_norm": [-0.6525943279266357, -0.6173503398895264], "gt_se_raw": [304.0, 306.0], "gt_se_norm": [0.5305987596511841, 0.5406684875488281], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003387.pt"}
{"idx": 3389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qMzt3yQFT-Q.mp4", "duration": 397.23016666666666, "prompt_text": "Please find the visual event described by the sentence 'When do the three men start clapping their hands and chanting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 250.0 - 259.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the three men start clapping their hands and chanting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 250.0 - 259.0 seconds.<|im_end|>\n", "base_se_raw": [250.0, 259.0], "base_se_norm": [0.2587161064147949, 0.30402982234954834], "gt_se_raw": [249.0, 259.0], "gt_se_norm": [0.2536811828613281, 0.30402982234954834], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003389.pt"}
{"idx": 3391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZUJJdeByJa0.mp4", "duration": 182.28, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of adding all the ingredients to the food processor before blending.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of adding all the ingredients to the food processor before blending.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 74.0], "base_se_norm": [-0.7915295362472534, -0.18806231021881104], "gt_se_raw": [19.0, 74.0], "gt_se_norm": [-0.7915295362472534, -0.18806231021881104], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003391.pt"}
{"idx": 3393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZUJJdeByJa0.mp4", "duration": 182.28, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the chef is rubbing the marinade all over the chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the chef is rubbing the marinade all over the chicken.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 133.0], "base_se_norm": [0.11915731430053711, 0.4592933654785156], "gt_se_raw": [101.0, 131.0], "gt_se_norm": [0.10818517208099365, 0.43734920024871826], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003393.pt"}
{"idx": 3395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wJ-mdxp_TY0.mp4", "duration": 62.77104166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman putting the flowers into the vase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman putting the flowers into the vase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8725526928901672], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9681381583213806, -0.8725526928901672], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003395.pt"}
{"idx": 3397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wJ-mdxp_TY0.mp4", "duration": 62.77104166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the woman puts her hand on her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the woman puts her hand on her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.10786885023117065, 0.05144023895263672], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.10786885023117065, 0.01957845687866211], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003397.pt"}
{"idx": 3399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wJ-mdxp_TY0.mp4", "duration": 62.77104166666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a glowing orb appears on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a glowing orb appears on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [0.08330202102661133, 0.1470257043838501], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [0.08330202102661133, 0.1470257043838501], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003399.pt"}
{"idx": 3401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_dGXDRZEkc.mp4", "duration": 25.025000000000002, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when an adult says 'really good girl'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when an adult says 'really good girl'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [0.5184814929962158, 0.7582417726516724], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.27872133255004883, 0.43856143951416016], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003401.pt"}
{"idx": 3403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W_dGXDRZEkc.mp4", "duration": 25.025000000000002, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a person's hands reach in from the left to adjust the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a person's hands reach in from the left to adjust the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.6803196668624878, -0.2807192802429199], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.838161826133728, 0.9980020523071289], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003403.pt"}
{"idx": 3405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RtZZZmmdgU.mp4", "duration": 146.07926666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the man remove his white tank top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man remove his white tank top?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.6029553413391113, -0.47973453998565674], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.5481904745101929, -0.47973453998565674], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003405.pt"}
{"idx": 3407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RtZZZmmdgU.mp4", "duration": 146.07926666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman in the striped shirt pulls out a stick for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman in the striped shirt pulls out a stick for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.288057804107666, -0.23329299688339233], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.2606753706932068, -0.23329299688339233], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003407.pt"}
{"idx": 3409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RtZZZmmdgU.mp4", "duration": 146.07926666666668, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the ULTA $20,000 banner is displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the ULTA $20,000 banner is displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 140.0], "base_se_norm": [0.8620027303695679, 0.9167674779891968], "gt_se_raw": [136.0, 139.0], "gt_se_norm": [0.8620027303695679, 0.9030762910842896], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003409.pt"}
{"idx": 3411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bwzLiQZDw2I.mp4", "duration": 141.18270833333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Elsa conjures magical ice diamonds that float around her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Elsa conjures magical ice diamonds that float around her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 40.0], "base_se_norm": [-0.4758564829826355, -0.433358371257782], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [-0.44752442836761475, -0.4191923141479492], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003411.pt"}
{"idx": 3413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bwzLiQZDw2I.mp4", "duration": 141.18270833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a small blue lizard sits on Elsa's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a small blue lizard sits on Elsa's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 93.0], "base_se_norm": [0.2749437093734741, 0.31744182109832764], "gt_se_raw": [90.0, 92.0], "gt_se_norm": [0.2749437093734741, 0.3032757043838501], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003413.pt"}
{"idx": 3415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IASj8rqh24c.mp4", "duration": 140.10471204188482, "prompt_text": "Please find the visual event described by the sentence 'When does the woman say she needs eight hours of sleep?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman say she needs eight hours of sleep?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.8572496175765991, -0.8144245147705078], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.8857997059822083, -0.82869952917099], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003415.pt"}
{"idx": 3417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IASj8rqh24c.mp4", "duration": 140.10471204188482, "prompt_text": "Please find the visual event described by the sentence 'When does the woman show her dog on the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman show her dog on the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 57.0], "base_se_norm": [-0.2434229850769043, -0.18632280826568604], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.22914797067642212, -0.18632280826568604], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003417.pt"}
{"idx": 3419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IASj8rqh24c.mp4", "duration": 140.10471204188482, "prompt_text": "Please find the visual event described by the sentence 'At what point does she talk about rewarding herself with food in the morning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she talk about rewarding herself with food in the morning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 72.0], "base_se_norm": [-0.043572425842285156, 0.027802705764770508], "gt_se_raw": [96.0, 102.0], "gt_se_norm": [0.3704036474227905, 0.45605385303497314], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003419.pt"}
{"idx": 3421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/faYA_upR_a0.mp4", "duration": 127.76, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man reveal the cake from the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man reveal the cake from the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.7338759899139404, -0.6869129538536072], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.7338759899139404, -0.7025673389434814], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003421.pt"}
{"idx": 3423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/faYA_upR_a0.mp4", "duration": 127.76, "prompt_text": "Please find the visual event described by the sentence 'Show me when the cat climbs into the empty box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the cat climbs into the empty box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 68.0], "base_se_norm": [-0.02943021059036255, 0.06449592113494873], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.017532825469970703, 0.06449592113494873], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003423.pt"}
{"idx": 3425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EAEu4brBayI.mp4", "duration": 47.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher hold up an open book to show the student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher hold up an open book to show the student?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.5795375108718872, -0.4533987045288086], "gt_se_raw": [11.0, 32.0], "gt_se_norm": [-0.5374912023544312, 0.3454800844192505], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003425.pt"}
{"idx": 3427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EAEu4brBayI.mp4", "duration": 47.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time does the boy raise his hand to touch the side of the monitor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the boy raise his hand to touch the side of the monitor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.4954449534416199, -0.4113524556159973], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.4954449534416199, -0.4533987045288086], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003427.pt"}
{"idx": 3429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EAEu4brBayI.mp4", "duration": 47.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the teacher begins reading the story.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the teacher begins reading the story.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.5795375108718872], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9579537510871887, -0.7477225065231323], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003429.pt"}
{"idx": 3431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/813_zdNJYa4.mp4", "duration": 192.42556666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the English muffins placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the English muffins placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 73.0], "base_se_norm": [-0.5946484804153442, -0.24126505851745605], "gt_se_raw": [40.0, 74.0], "gt_se_norm": [-0.5842548608779907, -0.23087143898010254], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003431.pt"}
{"idx": 3433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/813_zdNJYa4.mp4", "duration": 192.42556666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is cheese being sprinkled on the muffins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is cheese being sprinkled on the muffins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 117.0], "base_se_norm": [-0.0229988694190979, 0.21605455875396729], "gt_se_raw": [94.0, 116.0], "gt_se_norm": [-0.0229988694190979, 0.20566093921661377], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003433.pt"}
{"idx": 3435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XHY2eXoCeSY.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see three men gathered around an electrical training board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three men gathered around an electrical training board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003435.pt"}
{"idx": 3437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XHY2eXoCeSY.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man with red hair and a white t-shirt working on wiring inside a wooden frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man with red hair and a white t-shirt working on wiring inside a wooden frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003437.pt"}
{"idx": 3439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XHY2eXoCeSY.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the 'CAREERS' section of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the 'CAREERS' section of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003439.pt"}
{"idx": 3441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d6g1c18Cy-8.mp4", "duration": 211.97843333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the grey shirt place items into the white partitioned tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the grey shirt place items into the white partitioned tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 61.0], "base_se_norm": [-0.509384036064148, -0.4244697690010071], "gt_se_raw": [52.0, 62.0], "gt_se_norm": [-0.509384036064148, -0.41503483057022095], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003441.pt"}
{"idx": 3443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d6g1c18Cy-8.mp4", "duration": 211.97843333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the grey shirt put the black box away on the shelf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the grey shirt put the black box away on the shelf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 116.0], "base_se_norm": [-0.028203070163726807, 0.0944509506225586], "gt_se_raw": [101.0, 107.0], "gt_se_norm": [-0.0470728874206543, 0.00953662395477295], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003443.pt"}
{"idx": 3445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7AlNZmmVU7U.mp4", "duration": 63.797066666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the woman is sitting at the table and speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the woman is sitting at the table and speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 63.0], "base_se_norm": [-1.0, 0.9750124216079712], "gt_se_raw": [1.0, 63.0], "gt_se_norm": [-0.9686505794525146, 0.9750124216079712], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003445.pt"}
{"idx": 3447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7AlNZmmVU7U.mp4", "duration": 63.797066666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman describe her goal of distributing computers to the visually impaired?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman describe her goal of distributing computers to the visually impaired?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9686505794525146, -0.5924577713012695], "gt_se_raw": [31.0, 55.0], "gt_se_norm": [-0.02816849946975708, 0.7242171764373779], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003447.pt"}
{"idx": 3449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7AlNZmmVU7U.mp4", "duration": 63.797066666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman promises to revolutionize people's lives if she wins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman promises to revolutionize people's lives if she wins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [0.003180980682373047, 0.2539762258529663], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [0.7869160175323486, 0.9750124216079712], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003449.pt"}
{"idx": 3451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GHDz-XDD8OU.mp4", "duration": 30.932000000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the man sleeping on the couch with a bag of chips on his stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man sleeping on the couch with a bag of chips on his stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 18.0], "base_se_norm": [-0.2887624502182007, 0.1638432741165161], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.2887624502182007, -0.03013056516647339], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003451.pt"}
{"idx": 3453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GHDz-XDD8OU.mp4", "duration": 30.932000000000002, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the intruder eats the chip he just took.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the intruder eats the chip he just took.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [0.5517910718917847, 0.745764970779419], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.5517910718917847, 0.6811070442199707], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003453.pt"}
{"idx": 3455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TsujuQbDpI0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the employment rate for disabled Americans in 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the employment rate for disabled Americans in 1991?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9366739988327026], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003455.pt"}
{"idx": 3457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TsujuQbDpI0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When are the data for the year 2010 displayed on the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the data for the year 2010 displayed on the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9366739988327026], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003457.pt"}
{"idx": 3459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TsujuQbDpI0.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator mention that people with disabilities are employed at nearly half the rate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator mention that people with disabilities are employed at nearly half the rate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9031662940979004, -0.6126651763916016], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.2253304123878479, 0.355671763420105], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003459.pt"}
{"idx": 3461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D01ihkEgdho.mp4", "duration": 128.89556222889556, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a team is working on installing a window from the outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a team is working on installing a window from the outside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 19.0], "base_se_norm": [-0.7517369985580444, -0.7051876783370972], "gt_se_raw": [17.0, 18.0], "gt_se_norm": [-0.7362205982208252, -0.7207041382789612], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003461.pt"}
{"idx": 3463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D01ihkEgdho.mp4", "duration": 128.89556222889556, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when two women are examining a wall socket during renovation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when two women are examining a wall socket during renovation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 83.0], "base_se_norm": [0.13269996643066406, 0.2878643274307251], "gt_se_raw": [73.0, 83.0], "gt_se_norm": [0.13269996643066406, 0.2878643274307251], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003463.pt"}
{"idx": 3465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtY40NIWg0o.mp4", "duration": 130.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the woman talk about her experience at the biggest UC campus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman talk about her experience at the biggest UC campus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 48.0], "base_se_norm": [-0.49256789684295654, -0.2619169354438782], "gt_se_raw": [15.0, 26.0], "gt_se_norm": [-0.7693490386009216, -0.6002050638198853], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003465.pt"}
{"idx": 3467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtY40NIWg0o.mp4", "duration": 130.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the time when she talks about studying abroad in England with her guide dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when she talks about studying abroad in England with her guide dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 50.0], "base_se_norm": [-0.49256789684295654, -0.23116350173950195], "gt_se_raw": [49.0, 56.0], "gt_se_norm": [-0.24654024839401245, -0.13890314102172852], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003467.pt"}
{"idx": 3469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RtY40NIWg0o.mp4", "duration": 130.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where she says she had a great O&M instructor in high school.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where she says she had a great O&M instructor in high school.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.507944643497467, -0.3849307894706726], "gt_se_raw": [79.0, 84.0], "gt_se_norm": [0.2147616147994995, 0.2916452884674072], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003469.pt"}
{"idx": 3471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fmnUwvZAMjQ.mp4", "duration": 181.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does a dog appear and say 'Let's practice'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a dog appear and say 'Let's practice'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 46.0], "base_se_norm": [-0.5138121843338013, -0.49171268939971924], "gt_se_raw": [44.0, 46.0], "gt_se_norm": [-0.5138121843338013, -0.49171268939971924], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003471.pt"}
{"idx": 3473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fmnUwvZAMjQ.mp4", "duration": 181.0, "prompt_text": "Please find the visual event described by the sentence 'When are the three main ideas of the practice thesis listed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three main ideas of the practice thesis listed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 111.0], "base_se_norm": [0.14917123317718506, 0.22651934623718262], "gt_se_raw": [104.0, 121.0], "gt_se_norm": [0.14917123317718506, 0.33701658248901367], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003473.pt"}
{"idx": 3475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xNrR4JMACMg.mp4", "duration": 149.9498, "prompt_text": "Please find the visual event described by the sentence 'When does the presenter introduce the main topic of the presentation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the presenter introduce the main topic of the presentation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 21.0], "base_se_norm": [-1.0, -0.7199062705039978], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9866622090339661, -0.9466488361358643], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003475.pt"}
{"idx": 3477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xNrR4JMACMg.mp4", "duration": 149.9498, "prompt_text": "Please find the visual event described by the sentence 'At what point is a presentation slide detailing statistics about disability in Australia displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a presentation slide detailing statistics about disability in Australia displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 52.0], "base_se_norm": [-0.6398794651031494, -0.3064345717430115], "gt_se_raw": [27.0, 43.0], "gt_se_norm": [-0.6398794651031494, -0.42647475004196167], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003477.pt"}
{"idx": 3479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xNrR4JMACMg.mp4", "duration": 149.9498, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker summarize the message that movies about disability can convey?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker summarize the message that movies about disability can convey?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 149.0], "base_se_norm": [0.5071710348129272, 0.987331748008728], "gt_se_raw": [113.0, 124.0], "gt_se_norm": [0.5071710348129272, 0.6538867950439453], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003479.pt"}
{"idx": 3481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTxn2BuqyzU.mp4", "duration": 176.04166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy pulls a long vine out of his nose in the bathroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy pulls a long vine out of his nose in the bathroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 50.0], "base_se_norm": [-0.5114793181419373, -0.43195265531539917], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [-0.5001183748245239, -0.46603554487228394], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003481.pt"}
{"idx": 3483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTxn2BuqyzU.mp4", "duration": 176.04166666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment a nurse approaches the boy with a chainsaw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment a nurse approaches the boy with a chainsaw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 104.0], "base_se_norm": [0.13609468936920166, 0.18153846263885498], "gt_se_raw": [100.0, 104.0], "gt_se_norm": [0.13609468936920166, 0.18153846263885498], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003483.pt"}
{"idx": 3485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J_dBkatloPk.mp4", "duration": 14.633999999999999, "prompt_text": "Please find the visual event described by the sentence 'When can we see the crowd cheering and clapping in unison?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the crowd cheering and clapping in unison?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.9133524894714355], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, 0.7766844034194946], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003485.pt"}
{"idx": 3487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J_dBkatloPk.mp4", "duration": 14.633999999999999, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the '117' section sign is clearly visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the '117' section sign is clearly visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.9133524894714355], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.7266639471054077, -0.3166598081588745], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003487.pt"}
{"idx": 3489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J_dBkatloPk.mp4", "duration": 14.633999999999999, "prompt_text": "Please find the visual event described by the sentence 'Show me when the camera focuses on the fans sitting in the lower rows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the camera focuses on the fans sitting in the lower rows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.17999178171157837, 0.366680383682251], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [0.23001229763031006, 0.7766844034194946], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003489.pt"}
{"idx": 3491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qwGfbIw206k.mp4", "duration": 47.48076666666667, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the text 'Using favorite and familiar bright colored pants'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the text 'Using favorite and familiar bright colored pants'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 37.0], "base_se_norm": [-0.15755361318588257, 0.558525800704956], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.11543130874633789, 0.09518027305603027], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003491.pt"}
{"idx": 3493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qwGfbIw206k.mp4", "duration": 47.48076666666667, "prompt_text": "Please find the visual event described by the sentence 'The child successfully pulls the pants from the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'The child successfully pulls the pants from the couch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.5366544723510742, -0.32604289054870605], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [0.6848927736282349, 0.7691373825073242], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003493.pt"}
{"idx": 3495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVTOIJ4Q1DA.mp4", "duration": 97.66423333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the floral cardigan show a piece of tested fabric to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the floral cardigan show a piece of tested fabric to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 57.0], "base_se_norm": [-0.42660683393478394, 0.16726458072662354], "gt_se_raw": [28.0, 38.0], "gt_se_norm": [-0.42660683393478394, -0.2218235731124878], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003495.pt"}
{"idx": 3497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVTOIJ4Q1DA.mp4", "duration": 97.66423333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman in the floral top explains the purpose of the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman in the floral top explains the purpose of the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 27.0], "base_se_norm": [-0.9590433239936829, -0.4470852017402649], "gt_se_raw": [0.0, 22.0], "gt_se_norm": [-1.0, -0.5494768023490906], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003497.pt"}
{"idx": 3499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVTOIJ4Q1DA.mp4", "duration": 97.66423333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a blue top and a white scarf enter the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a blue top and a white scarf enter the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 74.0], "base_se_norm": [0.37204790115356445, 0.5153961181640625], "gt_se_raw": [68.0, 75.0], "gt_se_norm": [0.39252614974975586, 0.5358744859695435], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003499.pt"}
{"idx": 3501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XTW3LVp4pWA.mp4", "duration": 148.12, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a girl standing in front of a large stone building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a girl standing in front of a large stone building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.864974319934845, -0.837969183921814], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.864974319934845, -0.837969183921814], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003501.pt"}
{"idx": 3503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XTW3LVp4pWA.mp4", "duration": 148.12, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the doors for the Male and Female toilets are displayed side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the doors for the Male and Female toilets are displayed side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [-0.16284090280532837, -0.10883063077926636], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [-0.14933836460113525, -0.10883063077926636], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003503.pt"}
{"idx": 3505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_gNYItV44A.mp4", "duration": 123.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part of the playground with a sandy area and a purple fish-shaped spring rider.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part of the playground with a sandy area and a purple fish-shaped spring rider.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 116.0], "base_se_norm": [0.799513578414917, 0.8805727958679199], "gt_se_raw": [111.0, 117.0], "gt_se_norm": [0.799513578414917, 0.8967846632003784], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003505.pt"}
{"idx": 3507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_gNYItV44A.mp4", "duration": 123.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a person is riding a bike in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a person is riding a bike in the background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.7730343341827393, -0.7081869840621948], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.7730343341827393, -0.740610659122467], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003507.pt"}
{"idx": 3509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g_gNYItV44A.mp4", "duration": 123.36666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the shot of the merry-go-round with orange and purple railings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the shot of the merry-go-round with orange and purple railings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 100.0], "base_se_norm": [0.3455822467803955, 0.6211833953857422], "gt_se_raw": [83.0, 100.0], "gt_se_norm": [0.3455822467803955, 0.6211833953857422], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003509.pt"}
{"idx": 3511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RF99-5G-Hrk.mp4", "duration": 102.73596666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point are the two women looking at the gray jacket together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the two women looking at the gray jacket together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 37.0], "base_se_norm": [-0.5522502660751343, -0.27970701456069946], "gt_se_raw": [24.0, 31.0], "gt_se_norm": [-0.5327829122543335, -0.3965112566947937], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003511.pt"}
{"idx": 3513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RF99-5G-Hrk.mp4", "duration": 102.73596666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when a woman is showing off a purple coat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when a woman is showing off a purple coat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 68.0], "base_se_norm": [0.12910795211791992, 0.32378172874450684], "gt_se_raw": [57.0, 65.0], "gt_se_norm": [0.10964059829711914, 0.26537954807281494], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003513.pt"}
{"idx": 3515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KwOdedfBqHE.mp4", "duration": 196.82162499999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a long, open road stretching through a grassy plain under a partly cloudy sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a long, open road stretching through a grassy plain under a partly cloudy sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.7459628582000732, -0.6849939823150635], "gt_se_raw": [25.0, 31.0], "gt_se_norm": [-0.7459628582000732, -0.6849939823150635], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003515.pt"}
{"idx": 3517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KwOdedfBqHE.mp4", "duration": 196.82162499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point are the yellow flowers shown up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the yellow flowers shown up close?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [-0.40047240257263184, -0.3598264455795288], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.40047240257263184, -0.3598264455795288], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003517.pt"}
{"idx": 3519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KwOdedfBqHE.mp4", "duration": 196.82162499999998, "prompt_text": "Please find the visual event described by the sentence 'When is the seal of the state of Montana displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 183.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the seal of the state of Montana displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 183.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [183.0, 189.0], "base_se_norm": [0.8595517873764038, 0.9205206632614136], "gt_se_raw": [184.0, 189.0], "gt_se_norm": [0.869713306427002, 0.9205206632614136], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003519.pt"}
{"idx": 3521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FY0f8PYebKo.mp4", "duration": 123.6235, "prompt_text": "Please find the visual event described by the sentence 'Show the man in the red shirt and black jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man in the red shirt and black jacket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 30.0], "base_se_norm": [-0.7411495447158813, -0.514655351638794], "gt_se_raw": [17.0, 30.0], "gt_se_norm": [-0.7249714136123657, -0.514655351638794], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003521.pt"}
{"idx": 3523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FY0f8PYebKo.mp4", "duration": 123.6235, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman in a black tank top is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman in a black tank top is being interviewed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 51.0], "base_se_norm": [-0.3852301239967346, -0.17491412162780762], "gt_se_raw": [38.0, 51.0], "gt_se_norm": [-0.3852301239967346, -0.17491412162780762], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003523.pt"}
{"idx": 3525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iGtZnKy7-jM.mp4", "duration": 19.4194, "prompt_text": "Please find the visual event described by the sentence 'Show me when two boys are playing the electronic drums together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when two boys are playing the electronic drums together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, 0.9568061828613281], "gt_se_raw": [0.0, 19.0], "gt_se_norm": [-1.0, 0.9568061828613281], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003525.pt"}
{"idx": 3527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iGtZnKy7-jM.mp4", "duration": 19.4194, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person on the right put their hand on the boy in the middle to guide him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person on the right put their hand on the boy in the middle to guide him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 19.0], "base_se_norm": [-0.588040828704834, 0.9568061828613281], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.588040828704834, -0.4850510358810425], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003527.pt"}
{"idx": 3529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iGtZnKy7-jM.mp4", "duration": 19.4194, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time frame where the boy on the left hits the center drum pad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time frame where the boy on the left hits the center drum pad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [0.02989792823791504, 0.23587751388549805], "gt_se_raw": [16.0, 17.0], "gt_se_norm": [0.6478368043899536, 0.7508265972137451], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003529.pt"}
{"idx": 3531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZjejoT1gFOc.mp4", "duration": 198.63176666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a painting of early humans hunting a mammoth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a painting of early humans hunting a mammoth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.9395866990089417, -0.9093800783157349], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.9395866990089417, -0.9093800783157349], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003531.pt"}
{"idx": 3533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZjejoT1gFOc.mp4", "duration": 198.63176666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the video show people in white suits examining cave paintings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show people in white suits examining cave paintings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 134.0], "base_se_norm": [0.29888594150543213, 0.34923040866851807], "gt_se_raw": [128.0, 134.0], "gt_se_norm": [0.28881704807281494, 0.34923040866851807], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003533.pt"}
{"idx": 3535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_vUG5rqC6qI.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two men are tightening bolts on a red engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two men are tightening bolts on a red engine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003535.pt"}
{"idx": 3537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_vUG5rqC6qI.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man lying on his back under a vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man lying on his back under a vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08288431167602539, 0.3756735324859619], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003537.pt"}
{"idx": 3539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_vUG5rqC6qI.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we see three mechanics looking up at the underside of a car on a lift?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three mechanics looking up at the underside of a car on a lift?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [0.3756735324859619, 0.8342313766479492], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003539.pt"}
{"idx": 3541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b33M-dJu5gM.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the child use the device to say 'I want a spinning toy'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the child use the device to say 'I want a spinning toy'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 41.0], "base_se_norm": [-0.06542056798934937, 0.5327103137969971], "gt_se_raw": [32.0, 41.0], "gt_se_norm": [0.19626164436340332, 0.5327103137969971], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003541.pt"}
{"idx": 3543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b33M-dJu5gM.mp4", "duration": 53.5, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the child receives the blue spinning toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the child receives the blue spinning toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 47.0], "base_se_norm": [0.6822429895401001, 0.7570093870162964], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [0.6822429895401001, 0.8317756652832031], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003543.pt"}
{"idx": 3545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AB2oAgjjt3g.mp4", "duration": 62.1621, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the striped shirt introduce herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the striped shirt introduce herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.9356521368026733, -0.646086573600769], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.9356521368026733, -0.8069563508033752], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003545.pt"}
{"idx": 3547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AB2oAgjjt3g.mp4", "duration": 62.1621, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a purple shirt gesturing while explaining something?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a purple shirt gesturing while explaining something?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.16347742080688477, 0.029566168785095215], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.16347742080688477, -0.00260770320892334], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003547.pt"}
{"idx": 3549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AB2oAgjjt3g.mp4", "duration": 62.1621, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a woman in a black shirt is talking to a man with glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a woman in a black shirt is talking to a man with glasses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 53.0], "base_se_norm": [0.6408711671829224, 0.705219030380249], "gt_se_raw": [50.0, 52.0], "gt_se_norm": [0.6086971759796143, 0.6730451583862305], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003549.pt"}
{"idx": 3551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uDTB-0LqDw8.mp4", "duration": 152.5, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman is introducing the rose garden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman is introducing the rose garden.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 30.0], "base_se_norm": [-1.0, -0.6065573692321777], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9868852496147156, -0.8950819969177246], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003551.pt"}
{"idx": 3553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uDTB-0LqDw8.mp4", "duration": 152.5, "prompt_text": "Please find the visual event described by the sentence 'At what point does the monarch butterfly appear on the milkweed plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the monarch butterfly appear on the milkweed plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 62.0], "base_se_norm": [-0.3049180507659912, -0.18688523769378662], "gt_se_raw": [53.0, 63.0], "gt_se_norm": [-0.3049180507659912, -0.1737704873085022], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003553.pt"}
{"idx": 3555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s6Mk7Y3Q5QE.mp4", "duration": 38.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the boy operating the machine with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy operating the machine with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.3258426785469055], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.7925670146942139], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003555.pt"}
{"idx": 3557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s6Mk7Y3Q5QE.mp4", "duration": 38.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where another person's hand points to the paper in the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where another person's hand points to the paper in the machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.1927398443222046, 0.29645633697509766], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.1927398443222046, 0.29645633697509766], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003557.pt"}
{"idx": 3559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s6Mk7Y3Q5QE.mp4", "duration": 38.56666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy look up and stick his tongue out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy look up and stick his tongue out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [0.5038894414901733, 0.7631807327270508], "gt_se_raw": [31.0, 33.0], "gt_se_norm": [0.6076059341430664, 0.7113224267959595], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003559.pt"}
{"idx": 3561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PV34pW-53Os.mp4", "duration": 110.90245833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the example of the correct way for Santa to interact with a child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the example of the correct way for Santa to interact with a child.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 34.0], "base_se_norm": [-0.5491533279418945, -0.3868485689163208], "gt_se_raw": [25.0, 34.0], "gt_se_norm": [-0.5491533279418945, -0.3868485689163208], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003561.pt"}
{"idx": 3563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PV34pW-53Os.mp4", "duration": 110.90245833333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the female elf in the striped hat impatiently says 'Next!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the female elf in the striped hat impatiently says 'Next!'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 39.0], "base_se_norm": [-0.3327469825744629, -0.2966792583465576], "gt_se_raw": [38.0, 40.0], "gt_se_norm": [-0.31471312046051025, -0.278645396232605], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003563.pt"}
{"idx": 3565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Opbfh4wNu7Q.mp4", "duration": 60.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the man with a tattooed arm appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man with a tattooed arm appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 23.0], "base_se_norm": [-0.2671756148338318, -0.2338653802871704], "gt_se_raw": [22.0, 23.0], "gt_se_norm": [-0.2671756148338318, -0.2338653802871704], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003565.pt"}
{"idx": 3567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Opbfh4wNu7Q.mp4", "duration": 60.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man is looking at a framed photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man is looking at a framed photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 38.0], "base_se_norm": [0.19916725158691406, 0.26578760147094727], "gt_se_raw": [36.0, 38.0], "gt_se_norm": [0.19916725158691406, 0.26578760147094727], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003567.pt"}
{"idx": 3569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Opbfh4wNu7Q.mp4", "duration": 60.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see the final 'Avengers Endgame' title card?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the final 'Avengers Endgame' title card?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 59.0], "base_se_norm": [0.7654405832290649, 0.9653018712997437], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [0.7987508773803711, 0.8986814022064209], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003569.pt"}
{"idx": 3571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KhKaTh-YQNE.mp4", "duration": 55.522133333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman walk away from the boy and out of the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman walk away from the boy and out of the scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.7838699817657471, -0.6397832632064819], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.7478482723236084, -0.603761613368988], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003571.pt"}
{"idx": 3573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KhKaTh-YQNE.mp4", "duration": 55.522133333333336, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman comes back carrying a stuffed animal and other toys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman comes back carrying a stuffed animal and other toys.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 46.0], "base_se_norm": [0.4768885374069214, 0.6569968461990356], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [0.4768885374069214, 0.5849535465240479], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003573.pt"}
{"idx": 3575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E07s5ZYygMg.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'When is the man wearing blue flower-shaped sunglasses enjoying the sun by himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man wearing blue flower-shaped sunglasses enjoying the sun by himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 26.0], "base_se_norm": [-0.8940754532814026, -0.7245961427688599], "gt_se_raw": [18.0, 26.0], "gt_se_norm": [-0.8093358278274536, -0.7245961427688599], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003575.pt"}
{"idx": 3577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E07s5ZYygMg.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a woman feeds the man a strawberry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a woman feeds the man a strawberry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [-0.12082618474960327, -0.09964126348495483], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [-0.1314186453819275, -0.11023372411727905], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003577.pt"}
{"idx": 3579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E07s5ZYygMg.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'When can we see a group of people running into the ocean at dusk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a group of people running into the ocean at dusk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 168.0], "base_se_norm": [0.7583476305007935, 0.7795325517654419], "gt_se_raw": [166.0, 169.0], "gt_se_norm": [0.7583476305007935, 0.7901250123977661], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003579.pt"}
{"idx": 3581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tEyDRWmQEKs.mp4", "duration": 133.3332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when one student is giving verbal feedback to his peer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when one student is giving verbal feedback to his peer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 78.0], "base_se_norm": [-0.29499930143356323, 0.17000114917755127], "gt_se_raw": [35.0, 94.0], "gt_se_norm": [-0.47499948740005493, 0.41000139713287354], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003581.pt"}
{"idx": 3583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tEyDRWmQEKs.mp4", "duration": 133.3332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher point to the charts on the orange wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher point to the charts on the orange wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 34.0], "base_se_norm": [-0.9549999833106995, -0.48999953269958496], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.6099996566772461, -0.5649995803833008], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003583.pt"}
{"idx": 3585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hseWMRV3lA8.mp4", "duration": 119.753, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the brown suit explain that he has rearranged the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the brown suit explain that he has rearranged the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 52.0], "base_se_norm": [-0.3319582939147949, -0.1315457820892334], "gt_se_raw": [30.0, 37.0], "gt_se_norm": [-0.4989687204360962, -0.3820614218711853], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003585.pt"}
{"idx": 3587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hseWMRV3lA8.mp4", "duration": 119.753, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the brown suit sing a song to explain the new keyboard layout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the brown suit sing a song to explain the new keyboard layout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 64.0], "base_se_norm": [-0.3319582939147949, 0.06886672973632812], "gt_se_raw": [58.0, 64.0], "gt_se_norm": [-0.03133946657180786, 0.06886672973632812], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003587.pt"}
{"idx": 3589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hseWMRV3lA8.mp4", "duration": 119.753, "prompt_text": "Please find the visual event described by the sentence 'What are the start and end times when the reporter says 'So long Captain' and signs off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What are the start and end times when the reporter says 'So long Captain' and signs off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 88.0], "base_se_norm": [0.41958868503570557, 0.46969175338745117], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.41958868503570557, 0.46969175338745117], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003589.pt"}
{"idx": 3591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/II5UsqP2JAk.mp4", "duration": 181.0809, "prompt_text": "Please find the visual event described by the sentence 'At what point does someone wearing a gas mask fall through the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does someone wearing a gas mask fall through the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 45.0], "base_se_norm": [-0.5361189842224121, -0.5029845833778381], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.5250741243362427, -0.5029845833778381], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003591.pt"}
{"idx": 3593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/II5UsqP2JAk.mp4", "duration": 181.0809, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person on horseback in a field with ruined skyscrapers in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person on horseback in a field with ruined skyscrapers in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 105.0], "base_se_norm": [0.12656831741333008, 0.15970265865325928], "gt_se_raw": [102.0, 105.0], "gt_se_norm": [0.12656831741333008, 0.15970265865325928], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003593.pt"}
{"idx": 3595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B-dn8jMxKeU.mp4", "duration": 44.67796666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the person in the hoodie look at the pink flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in the hoodie look at the pink flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.6418816447257996, -0.5523520708084106], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.6866464614868164, -0.5971168279647827], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003595.pt"}
{"idx": 3597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B-dn8jMxKeU.mp4", "duration": 44.67796666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man taking a photo on the beach with his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man taking a photo on the beach with his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.1942337155342102, -0.10470414161682129], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.1942337155342102, -0.10470414161682129], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003597.pt"}
{"idx": 3599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B-dn8jMxKeU.mp4", "duration": 44.67796666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the people walking through the airport?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the people walking through the airport?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 35.0], "base_se_norm": [0.47723817825317383, 0.5667678117752075], "gt_se_raw": [33.0, 34.0], "gt_se_norm": [0.47723817825317383, 0.5220030546188354], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003599.pt"}
{"idx": 3601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PcnqtNANWVw.mp4", "duration": 471.9957224969924, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a man is playing the violin on the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a man is playing the violin on the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.8008456230163574, -0.7627097368240356], "gt_se_raw": [48.0, 56.0], "gt_se_norm": [-0.7966083288192749, -0.7627097368240356], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003601.pt"}
{"idx": 3603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PcnqtNANWVw.mp4", "duration": 471.9957224969924, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a man and a woman are holding up illustrated signs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a man and a woman are holding up illustrated signs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 88.0], "base_se_norm": [-0.648301899433136, -0.6271152496337891], "gt_se_raw": [84.0, 88.0], "gt_se_norm": [-0.6440645456314087, -0.6271152496337891], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003603.pt"}
{"idx": 3605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PUtHH2rw7KM.mp4", "duration": 96.36293333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the white transmitter box shown on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the white transmitter box shown on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.7509415745735168, -0.6264123916625977], "gt_se_raw": [12.0, 18.0], "gt_se_norm": [-0.7509415745735168, -0.6264123916625977], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003605.pt"}
{"idx": 3607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PUtHH2rw7KM.mp4", "duration": 96.36293333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the person sweeps the receiver to the left, causing the audio to become staticky and stop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the person sweeps the receiver to the left, causing the audio to become staticky and stop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.30755674839019775, 0.41133105754852295], "gt_se_raw": [46.0, 52.0], "gt_se_norm": [-0.04527604579925537, 0.0792531967163086], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003607.pt"}
{"idx": 3609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PUtHH2rw7KM.mp4", "duration": 96.36293333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man quickly sweeps the receiver back and forth to find the center of the signal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man quickly sweeps the receiver back and forth to find the center of the signal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.5433928966522217, -0.4188636541366577], "gt_se_raw": [83.0, 88.0], "gt_se_norm": [0.7226541042327881, 0.8264285326004028], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003609.pt"}
{"idx": 3611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nhPG6tC_TKU.mp4", "duration": 50.38366666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does a person insert money into the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a person insert money into the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.5633505582809448, -0.48395973443984985], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.5633505582809448, -0.48395973443984985], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003611.pt"}
{"idx": 3613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nhPG6tC_TKU.mp4", "duration": 50.38366666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman takes the coconut from the dispensing slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman takes the coconut from the dispensing slot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [0.07177591323852539, 0.19086205959320068], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.11147129535675049, 0.19086205959320068], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003613.pt"}
{"idx": 3615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2_Z1KdMtDD4.mp4", "duration": 29.963266666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the train rounding the bend?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the train rounding the bend?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.13227081298828125], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.13227081298828125], "start_token_span": [62, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003615.pt"}
{"idx": 3617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2_Z1KdMtDD4.mp4", "duration": 29.963266666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman wearing a blue shirt and sunglasses is seen sitting on the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman wearing a blue shirt and sunglasses is seen sitting on the train.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [0.20147109031677246, 0.468464732170105], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [0.20147109031677246, 0.468464732170105], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003617.pt"}
{"idx": 3619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2_Z1KdMtDD4.mp4", "duration": 29.963266666666666, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the passenger cars are passing the railroad crossing signal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the passenger cars are passing the railroad crossing signal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 29.0], "base_se_norm": [0.5352131128311157, 0.9357035160064697], "gt_se_raw": [20.0, 28.0], "gt_se_norm": [0.3349679708480835, 0.868955135345459], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003619.pt"}
{"idx": 3621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZLZOnqxa99c.mp4", "duration": 55.180125, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot where a dolphin opens its eye.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot where a dolphin opens its eye.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.34759122133255005, -0.23885637521743774], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.38383615016937256, -0.23885637521743774], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003621.pt"}
{"idx": 3623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZLZOnqxa99c.mp4", "duration": 55.180125, "prompt_text": "Please find the visual event described by the sentence 'At what point is there an aerial view of the dolphins swimming together over the reef?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is there an aerial view of the dolphins swimming together over the reef?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 53.0], "base_se_norm": [0.014858126640319824, 0.9209815263748169], "gt_se_raw": [28.0, 39.0], "gt_se_norm": [0.014858126640319824, 0.4135524034500122], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003623.pt"}
{"idx": 3625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gm5ItoIJ4sg.mp4", "duration": 49.5495, "prompt_text": "Please find the visual event described by the sentence 'When are the shoelaces crossed to form the first knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the shoelaces crossed to form the first knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 39.0], "base_se_norm": [0.08981931209564209, 0.574183464050293], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.6367269158363342, -0.43490850925445557], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003625.pt"}
{"idx": 3627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gm5ItoIJ4sg.mp4", "duration": 49.5495, "prompt_text": "Please find the visual event described by the sentence 'At what point is the first lace pushed through the circle to create a loop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the first lace pushed through the circle to create a loop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [0.08981931209564209, 0.37236499786376953], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [0.08981931209564209, 0.37236499786376953], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003627.pt"}
{"idx": 3629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gm5ItoIJ4sg.mp4", "duration": 49.5495, "prompt_text": "Please find the visual event described by the sentence 'When are the two loops pulled to tighten the final bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two loops pulled to tighten the final bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 46.0], "base_se_norm": [0.6952744722366333, 0.8567291498184204], "gt_se_raw": [42.0, 47.0], "gt_se_norm": [0.6952744722366333, 0.8970928192138672], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003629.pt"}
{"idx": 3631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wvaG-frUcV4.mp4", "duration": 11.866666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is an adult guiding a child's hand to make a gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an adult guiding a child's hand to make a gesture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, 0.685393214225769], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.6629213094711304, 0.17977523803710938], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003631.pt"}
{"idx": 3633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wvaG-frUcV4.mp4", "duration": 11.866666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the brand name 'PlayCore' can be seen on the yellow equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the brand name 'PlayCore' can be seen on the yellow equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.49438202381134033, 0.685393214225769], "gt_se_raw": [3.0, 9.0], "gt_se_norm": [-0.49438202381134033, 0.516853928565979], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003633.pt"}
{"idx": 3635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x5U8zODIz9s.mp4", "duration": 142.20873333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the person type the word 'fix'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person type the word 'fix'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 102.0], "base_se_norm": [0.293872594833374, 0.4345109462738037], "gt_se_raw": [34.0, 42.0], "gt_se_norm": [-0.5218296647071838, -0.4093189835548401], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003635.pt"}
{"idx": 3637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x5U8zODIz9s.mp4", "duration": 142.20873333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person type the word 'build'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person type the word 'build'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.7468509674072266, -0.6905956268310547], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [-0.043659329414367676, -0.0014678239822387695], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003637.pt"}
{"idx": 3639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x5U8zODIz9s.mp4", "duration": 142.20873333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the device say 'Build train remote control'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the device say 'Build train remote control'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.8734254837036133, -0.8312339782714844], "gt_se_raw": [130.0, 133.0], "gt_se_norm": [0.8282983303070068, 0.8704898357391357], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003639.pt"}
{"idx": 3641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q-SGjT4CPtQ.mp4", "duration": 56.51479166666666, "prompt_text": "Please find the visual event described by the sentence 'When are the emoji symbols displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the emoji symbols displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.3629986643791199, -0.2568317651748657], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.3629986643791199, -0.2568317651748657], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003641.pt"}
{"idx": 3643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q-SGjT4CPtQ.mp4", "duration": 56.51479166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video look up at the sun through the tree leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video look up at the sun through the tree leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 32.0], "base_se_norm": [-0.22144275903701782, 0.13244688510894775], "gt_se_raw": [22.0, 32.0], "gt_se_norm": [-0.22144275903701782, 0.13244688510894775], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003643.pt"}
{"idx": 3645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1zyNbdDgUb8.mp4", "duration": 466.3825833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the kitten getting a bath?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the kitten getting a bath?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9914233684539795, -0.978558361530304], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9871350526809692, -0.978558361530304], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003645.pt"}
{"idx": 3647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1zyNbdDgUb8.mp4", "duration": 466.3825833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman engaged in a video call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman engaged in a video call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 42.0], "base_se_norm": [-0.8584852814674377, -0.8198903799057007], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.8584852814674377, -0.8456202745437622], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003647.pt"}
{"idx": 3649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1zyNbdDgUb8.mp4", "duration": 466.3825833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cat is licking its paw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 414.0 - 417.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cat is licking its paw.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 414.0 - 417.0 seconds.<|im_end|>\n", "base_se_raw": [414.0, 417.0], "base_se_norm": [0.7753665447235107, 0.7882314920425415], "gt_se_raw": [414.0, 416.0], "gt_se_norm": [0.7753665447235107, 0.7839431762695312], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003649.pt"}
{"idx": 3651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DAHbtsjuNws.mp4", "duration": 132.96616666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man playing the clarinet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man playing the clarinet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.5938816666603088, -0.5186745524406433], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.5938816666603088, -0.5186745524406433], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003651.pt"}
{"idx": 3653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DAHbtsjuNws.mp4", "duration": 132.96616666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where a woman in a wheelchair is dancing with two other women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where a woman in a wheelchair is dancing with two other women.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 98.0], "base_se_norm": [0.06794083118438721, 0.4740591049194336], "gt_se_raw": [70.0, 85.0], "gt_se_norm": [0.05289936065673828, 0.27852070331573486], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003653.pt"}
{"idx": 3655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/emx92kBKads.mp4", "duration": 315.76666666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the video display information about China's potential to become the leading English-speaking country?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display information about China's potential to become the leading English-speaking country?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.8163200616836548, -0.7909848690032959], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.8163200616836548, -0.7909848690032959], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003655.pt"}
{"idx": 3657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/emx92kBKads.mp4", "duration": 315.76666666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video mention that one out of eight couples married in the U.S. met online?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video mention that one out of eight couples married in the U.S. met online?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 89.0], "base_se_norm": [-0.45529401302337646, -0.4362925887107849], "gt_se_raw": [85.0, 89.0], "gt_se_norm": [-0.4616277813911438, -0.4362925887107849], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003657.pt"}
{"idx": 3659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/emx92kBKads.mp4", "duration": 315.76666666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment that illustrates how long it took for the internet to reach a market audience of 50 million.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 174.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment that illustrates how long it took for the internet to reach a market audience of 50 million.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 174.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 174.0], "base_se_norm": [-0.024596214294433594, 0.10207962989807129], "gt_se_raw": [164.0, 167.0], "gt_se_norm": [0.03874170780181885, 0.057743072509765625], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003659.pt"}
{"idx": 3661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wKcPU0WPdqE.mp4", "duration": 89.82315648982316, "prompt_text": "Please find the visual event described by the sentence 'Show the low-angle shot of the runner's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the low-angle shot of the runner's legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 47.0], "base_se_norm": [0.0019687414169311523, 0.04650068283081055], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [0.0019687414169311523, 0.04650068283081055], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003661.pt"}
{"idx": 3663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wKcPU0WPdqE.mp4", "duration": 89.82315648982316, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot of the food pantry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot of the food pantry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [0.31369245052337646, 0.40275633335113525], "gt_se_raw": [59.0, 64.0], "gt_se_norm": [0.31369245052337646, 0.4250222444534302], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003663.pt"}
{"idx": 3665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aqXFcRRQZCM.mp4", "duration": 56.181124999999994, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman in the floral dress sitting with her arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman in the floral dress sitting with her arms crossed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.1812196969985962], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.893202543258667], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003665.pt"}
{"idx": 3667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aqXFcRRQZCM.mp4", "duration": 56.181124999999994, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman in a denim shirt is surprised after tasting the jelly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman in a denim shirt is surprised after tasting the jelly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.11002141237258911, -0.0032239556312561035], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.07442224025726318, -0.0032239556312561035], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003667.pt"}
{"idx": 3669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aqXFcRRQZCM.mp4", "duration": 56.181124999999994, "prompt_text": "Please find the visual event described by the sentence 'At what point does the text 'SUBSCRIBE TO DAD' show up on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the text 'SUBSCRIBE TO DAD' show up on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 55.0], "base_se_norm": [0.3527674674987793, 0.9579529762268066], "gt_se_raw": [38.0, 56.0], "gt_se_norm": [0.3527674674987793, 0.9935520887374878], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003669.pt"}
{"idx": 3671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wekSrZ-d1bM.mp4", "duration": 113.94728061394729, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a student is playing a xylophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a student is playing a xylophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 92.0], "base_se_norm": [0.2286384105682373, 0.6147818565368652], "gt_se_raw": [71.0, 74.0], "gt_se_norm": [0.24619030952453613, 0.2988462448120117], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003671.pt"}
{"idx": 3673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wekSrZ-d1bM.mp4", "duration": 113.94728061394729, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the students playing the flutes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the students playing the flutes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 42.0], "base_se_norm": [-0.38568079471588135, -0.2628169655799866], "gt_se_raw": [35.0, 42.0], "gt_se_norm": [-0.38568079471588135, -0.2628169655799866], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003673.pt"}
{"idx": 3675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YMpCN8CmZh8.mp4", "duration": 73.20646666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the man walking with the assistance of a woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man walking with the assistance of a woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 72.0], "base_se_norm": [-0.9726800322532654, 0.9670393466949463], "gt_se_raw": [1.0, 48.0], "gt_se_norm": [-0.9726800322532654, 0.3113595247268677], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003675.pt"}
{"idx": 3677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YMpCN8CmZh8.mp4", "duration": 73.20646666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman knock on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman knock on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 47.0], "base_se_norm": [0.1747596263885498, 0.28403961658477783], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [0.584559440612793, 0.6391994953155518], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003677.pt"}
{"idx": 3679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YMpCN8CmZh8.mp4", "duration": 73.20646666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two people moving forward in the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two people moving forward in the hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 40.0], "base_se_norm": [-0.9726800322532654, 0.09279966354370117], "gt_se_raw": [11.0, 27.0], "gt_se_norm": [-0.6994800567626953, -0.26236027479171753], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003679.pt"}
{"idx": 3681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdn2xstSvwo.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of someone using a measuring tool on a metal part.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of someone using a measuring tool on a metal part.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.47478586435317993, -0.2778306007385254], "gt_se_raw": [7.0, 11.0], "gt_se_norm": [-0.540437638759613, -0.2778306007385254], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003681.pt"}
{"idx": 3683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kdn2xstSvwo.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an instructor is teaching a class using a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an instructor is teaching a class using a whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.015223503112792969, 0.3786870241165161], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003683.pt"}
{"idx": 3685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODAfdbHBtGE.mp4", "duration": 37.971266666666665, "prompt_text": "Please find the visual event described by the sentence 'When does someone knock on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does someone knock on the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.8946571946144104], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9473285675048828, -0.8946571946144104], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003685.pt"}
{"idx": 3687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODAfdbHBtGE.mp4", "duration": 37.971266666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman mention making Christmas ornaments?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman mention making Christmas ornaments?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [-0.1572575569152832, 0.053428053855895996], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.4206145405769348, -0.3152717351913452], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003687.pt"}
{"idx": 3689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ODAfdbHBtGE.mp4", "duration": 37.971266666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman explicitly states that her family does not celebrate Christian holidays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman explicitly states that her family does not celebrate Christian holidays.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.05191469192504883, 0.15877091884613037], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [0.2641136646270752, 0.42212796211242676], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003689.pt"}
{"idx": 3691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H3DvNMAEWyM.mp4", "duration": 16.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does one woman draw a smiley face on the other woman's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does one woman draw a smiley face on the other woman's back?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [0.2072434425354004, 0.6901407241821289], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [0.3279677629470825, 0.6901407241821289], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003691.pt"}
{"idx": 3693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/H3DvNMAEWyM.mp4", "duration": 16.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman on the left is explaining how to visualize things.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman on the left is explaining how to visualize things.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, 0.08651900291442871], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.8792756795883179, -0.5171025991439819], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003693.pt"}
{"idx": 3695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qlvtkpn9Xn8.mp4", "duration": 127.375, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the blue shirt shown speaking alone to the camera at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the blue shirt shown speaking alone to the camera at the beginning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.8743866682052612], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.8743866682052612], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003695.pt"}
{"idx": 3697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qlvtkpn9Xn8.mp4", "duration": 127.375, "prompt_text": "Please find the visual event described by the sentence 'At what point does the patient mention she is feeling exhausted and that her house renovation is making her more tired?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the patient mention she is feeling exhausted and that her house renovation is making her more tired?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 58.0], "base_se_norm": [-0.26202160120010376, -0.08930325508117676], "gt_se_raw": [49.0, 62.0], "gt_se_norm": [-0.23061823844909668, -0.026496589183807373], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003697.pt"}
{"idx": 3699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qlvtkpn9Xn8.mp4", "duration": 127.375, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the doctor picks up the purple tray and moves his chair to a new position.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the doctor picks up the purple tray and moves his chair to a new position.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.6388616561889648, -0.5603532791137695], "gt_se_raw": [92.0, 107.0], "gt_se_norm": [0.4445534944534302, 0.6800785064697266], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003699.pt"}
{"idx": 3701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1OJ14e6EMQ.mp4", "duration": 7.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the boy sits up and crosses his legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the boy sits up and crosses his legs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.49579834938049316, 0.008403301239013672], "gt_se_raw": [2.0, 4.0], "gt_se_norm": [-0.49579834938049316, 0.008403301239013672], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003701.pt"}
{"idx": 3703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r1OJ14e6EMQ.mp4", "duration": 7.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy covers his face with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy covers his face with his hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 7.0], "base_se_norm": [0.26050424575805664, 0.7647058963775635], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [0.26050424575805664, 0.7647058963775635], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003703.pt"}
{"idx": 3705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQz8oCsHbY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Cortana show the weather forecast for tomorrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Cortana show the weather forecast for tomorrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.9420468807220459, -0.8840938210487366], "gt_se_raw": [7.0, 13.0], "gt_se_norm": [-0.9420468807220459, -0.8923728466033936], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003705.pt"}
{"idx": 3707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQz8oCsHbY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when Cortana tells a joke about a milk truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when Cortana tells a joke about a milk truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [-0.4370270371437073, -0.39563196897506714], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [-0.5115381479263306, -0.4784221053123474], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003707.pt"}
{"idx": 3709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQz8oCsHbY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the computer displays today's top headlines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the computer displays today's top headlines.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 172.0], "base_se_norm": [0.28324711322784424, 0.4239903688430786], "gt_se_raw": [155.0, 171.0], "gt_se_norm": [0.28324711322784424, 0.4157114028930664], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003709.pt"}
{"idx": 3711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VsJKPdOuzZw.mp4", "duration": 93.21812499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man has trouble drinking from the cup and spills the drink on his shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man has trouble drinking from the cup and spills the drink on his shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 92.0], "base_se_norm": [0.8451347351074219, 0.9738650321960449], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.8236796855926514, 0.8880448341369629], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003711.pt"}
{"idx": 3713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VsJKPdOuzZw.mp4", "duration": 93.21812499999999, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man is eating with a spoon and makes a strained face while looking up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man is eating with a spoon and makes a strained face while looking up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 55.0], "base_se_norm": [-0.09888768196105957, 0.18002808094024658], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.07743263244628906, -0.03452253341674805], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003713.pt"}
{"idx": 3715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T8NJLxTIEO0.mp4", "duration": 91.9919, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in the purple shirt using a magnifying device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in the purple shirt using a magnifying device.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9782589673995972, -0.9347768425941467], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9782589673995972, -0.9347768425941467], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003715.pt"}
{"idx": 3717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T8NJLxTIEO0.mp4", "duration": 91.9919, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man in a suit is giving a presentation with a microphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man in a suit is giving a presentation with a microphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.6738843321800232, -0.6086611747741699], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.6738843321800232, -0.6304022073745728], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003717.pt"}
{"idx": 3719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T8NJLxTIEO0.mp4", "duration": 91.9919, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the white shirt writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the white shirt writing on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 54.0], "base_se_norm": [0.13053429126739502, 0.17401647567749023], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [0.13053429126739502, 0.17401647567749023], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003719.pt"}
{"idx": 3721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3eMgzKC39Q.mp4", "duration": 196.70000000000002, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the message 'Blue Bed was dismantled by iTMG' appear in the chat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the message 'Blue Bed was dismantled by iTMG' appear in the chat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 82.0], "base_se_norm": [-0.18657851219177246, -0.16624301671981812], "gt_se_raw": [79.0, 81.0], "gt_se_norm": [-0.19674628973007202, -0.17641079425811768], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003721.pt"}
{"idx": 3723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3eMgzKC39Q.mp4", "duration": 196.70000000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the yellow player break the blocks to free the blue player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 150.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the yellow player break the blocks to free the blue player?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 150.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [150.0, 159.0], "base_se_norm": [0.5251652002334595, 0.6166751384735107], "gt_se_raw": [149.0, 157.0], "gt_se_norm": [0.5149974822998047, 0.5963395833969116], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003723.pt"}
{"idx": 3725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eTVqkWp7oCs.mp4", "duration": 57.265541666666664, "prompt_text": "Please find the visual event described by the sentence 'When are the red arrows moving between the groups of people on the map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the red arrows moving between the groups of people on the map?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 13.0], "base_se_norm": [-0.7904499173164368, -0.5459747910499573], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.7555248737335205, -0.6856748461723328], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003725.pt"}
{"idx": 3727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eTVqkWp7oCs.mp4", "duration": 57.265541666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the red toolbox appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the red toolbox appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 35.0], "base_se_norm": [-0.23164963722229004, 0.22237563133239746], "gt_se_raw": [22.0, 35.0], "gt_se_norm": [-0.23164963722229004, 0.22237563133239746], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003727.pt"}
{"idx": 3729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eTVqkWp7oCs.mp4", "duration": 57.265541666666664, "prompt_text": "Please find the visual event described by the sentence 'When is the list of words like 'Mentoring', 'Networking', and 'Jobs' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the list of words like 'Mentoring', 'Networking', and 'Jobs' shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 51.0], "base_se_norm": [0.5716257095336914, 0.7811758518218994], "gt_se_raw": [45.0, 52.0], "gt_se_norm": [0.5716257095336914, 0.8161008358001709], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003729.pt"}
{"idx": 3731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rdgoehu_0SE.mp4", "duration": 346.0790666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the purple and white checkered shirt speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the purple and white checkered shirt speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 73.0], "base_se_norm": [-0.7226067781448364, -0.5781310796737671], "gt_se_raw": [48.0, 73.0], "gt_se_norm": [-0.7226067781448364, -0.5781310796737671], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003731.pt"}
{"idx": 3733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rdgoehu_0SE.mp4", "duration": 346.0790666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the person in the pink shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person in the pink shirt speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 201.0], "base_se_norm": [-0.17937827110290527, 0.1615842580795288], "gt_se_raw": [142.0, 202.0], "gt_se_norm": [-0.17937827110290527, 0.16736328601837158], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003733.pt"}
{"idx": 3735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uE74-8YAV9E.mp4", "duration": 122.56, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video explain what the main idea is in writing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video explain what the main idea is in writing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 22.0], "base_se_norm": [-0.8368146419525146, -0.6409921646118164], "gt_se_raw": [10.0, 20.0], "gt_se_norm": [-0.8368146419525146, -0.6736292243003845], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003735.pt"}
{"idx": 3737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uE74-8YAV9E.mp4", "duration": 122.56, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the hierarchical relationship between main ideas and supporting details?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the hierarchical relationship between main ideas and supporting details?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 79.0], "base_se_norm": [-0.02088773250579834, 0.28916454315185547], "gt_se_raw": [61.0, 79.0], "gt_se_norm": [-0.0045691728591918945, 0.28916454315185547], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003737.pt"}
{"idx": 3739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uE74-8YAV9E.mp4", "duration": 122.56, "prompt_text": "Please find the visual event described by the sentence 'Find the part that lists transition words used to introduce major and minor details.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part that lists transition words used to introduce major and minor details.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 116.0], "base_se_norm": [0.6808093786239624, 0.8929504156112671], "gt_se_raw": [102.0, 116.0], "gt_se_norm": [0.6644909381866455, 0.8929504156112671], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003739.pt"}
{"idx": 3741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lO_ibzMt9MQ.mp4", "duration": 36.54496168133596, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman places the first piece of food on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman places the first piece of food on the counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.3980017900466919, -0.2885475754737854], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [-0.3980017900466919, -0.34327471256256104], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003741.pt"}
{"idx": 3743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lO_ibzMt9MQ.mp4", "duration": 36.54496168133596, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the woman explains that rats inhabit every continent except Antarctica.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the woman explains that rats inhabit every continent except Antarctica.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [0.47763192653656006, 0.7512674331665039], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.03981506824493408, 0.2039964199066162], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003743.pt"}
{"idx": 3745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BWYNt1N6xWQ.mp4", "duration": 259.7595, "prompt_text": "Please find the visual event described by the sentence 'When does the struggle begin after the boy touches the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the struggle begin after the boy touches the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 27.0], "base_se_norm": [-0.8845085501670837, -0.7921153903007507], "gt_se_raw": [14.0, 27.0], "gt_se_norm": [-0.8922079801559448, -0.7921153903007507], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003745.pt"}
{"idx": 3747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BWYNt1N6xWQ.mp4", "duration": 259.7595, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher start spelling the word 'cake'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher start spelling the word 'cake'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 68.0], "base_se_norm": [-0.6920228004455566, -0.4764387607574463], "gt_se_raw": [81.0, 88.0], "gt_se_norm": [-0.3763461709022522, -0.3224501609802246], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003747.pt"}
{"idx": 3749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BWYNt1N6xWQ.mp4", "duration": 259.7595, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the teacher hands the glass of milk to the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 218.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the teacher hands the glass of milk to the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 218.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [218.0, 222.0], "base_se_norm": [0.6784757375717163, 0.7092734575271606], "gt_se_raw": [218.0, 221.0], "gt_se_norm": [0.6784757375717163, 0.7015739679336548], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003749.pt"}
{"idx": 3751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lN05UfmIAh4.mp4", "duration": 61.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where the applicant touches the necklace on her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where the applicant touches the necklace on her chest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.3446919918060303, -0.24639582633972168], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.31192660331726074, -0.24639582633972168], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003751.pt"}
{"idx": 3753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lN05UfmIAh4.mp4", "duration": 61.04, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the people in the room laugh together in relief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the people in the room laugh together in relief?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 57.0], "base_se_norm": [0.7365661859512329, 0.867627739906311], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.7693315744400024, 0.8348623514175415], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003753.pt"}
{"idx": 3755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mv1FKi_-A1I.mp4", "duration": 138.27146666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the pinwheel is first successfully assembled and shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the pinwheel is first successfully assembled and shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 137.0], "base_se_norm": [0.7646446228027344, 0.9816091060638428], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.11767768859863281, -0.05982053279876709], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003755.pt"}
{"idx": 3757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mv1FKi_-A1I.mp4", "duration": 138.27146666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the two women are holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the two women are holding hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 12.0], "base_se_norm": [-0.9855356812477112, -0.8264284133911133], "gt_se_raw": [13.0, 21.0], "gt_se_norm": [-0.8119640946388245, -0.6962497234344482], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003757.pt"}
{"idx": 3759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mv1FKi_-A1I.mp4", "duration": 138.27146666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman begin attaching the yellow petals to the pinwheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman begin attaching the yellow petals to the pinwheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 122.0], "base_se_norm": [0.547680139541626, 0.7646446228027344], "gt_se_raw": [119.0, 128.0], "gt_se_norm": [0.7212517261505127, 0.8514304161071777], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003759.pt"}
{"idx": 3761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FldgzyEOSH8.mp4", "duration": 491.79130000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a drive-thru COVID-19 testing site is set up in a parking lot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a drive-thru COVID-19 testing site is set up in a parking lot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 86.0], "base_se_norm": [-0.6746587753295898, -0.6502581834793091], "gt_se_raw": [81.0, 87.0], "gt_se_norm": [-0.6705919504165649, -0.6461913585662842], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003761.pt"}
{"idx": 3763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FldgzyEOSH8.mp4", "duration": 491.79130000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where women are looking at a long table of auction items.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 290.0 - 311.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where women are looking at a long table of auction items.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 290.0 - 311.0 seconds.<|im_end|>\n", "base_se_raw": [290.0, 311.0], "base_se_norm": [0.17936205863952637, 0.2647641897201538], "gt_se_raw": [291.0, 305.0], "gt_se_norm": [0.18342888355255127, 0.24036359786987305], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003763.pt"}
{"idx": 3765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ynUw0YsrmSg.mp4", "duration": 218.13458333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the women holding a standard plank?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the women holding a standard plank?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 53.0], "base_se_norm": [-0.8166269659996033, -0.514061450958252], "gt_se_raw": [22.0, 52.0], "gt_se_norm": [-0.7982896566390991, -0.5232300758361816], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003765.pt"}
{"idx": 3767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ynUw0YsrmSg.mp4", "duration": 218.13458333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point are they doing a side plank on their right arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are they doing a side plank on their right arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 113.0], "base_se_norm": [-0.23900192975997925, 0.03605759143829346], "gt_se_raw": [62.0, 91.0], "gt_se_norm": [-0.43154358863830566, -0.16565269231796265], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003767.pt"}
{"idx": 3769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ynUw0YsrmSg.mp4", "duration": 218.13458333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see them doing Spiderman planks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 192.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see them doing Spiderman planks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 192.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 192.0], "base_se_norm": [0.5861767530441284, 0.7603811025619507], "gt_se_raw": [149.0, 179.0], "gt_se_norm": [0.3661290407180786, 0.6411886215209961], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003769.pt"}
{"idx": 3771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JNyn_w3hdZ0.mp4", "duration": 145.56208333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman in a striped jumpsuit is shouting into a megaphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman in a striped jumpsuit is shouting into a megaphone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.6702437996864319, -0.5603250861167908], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.6565039753913879, -0.6152844429016113], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003771.pt"}
{"idx": 3773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JNyn_w3hdZ0.mp4", "duration": 145.56208333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a young man in a white t-shirt and his family are posing for a group photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a young man in a white t-shirt and his family are posing for a group photo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.051950931549072266, 0.0030083656311035156], "gt_se_raw": [70.0, 73.0], "gt_se_norm": [-0.03821110725402832, 0.0030083656311035156], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003773.pt"}
{"idx": 3775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8bRkdkfWv3g.mp4", "duration": 110.92, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a ship travels from South America to Europe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a ship travels from South America to Europe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 56.0], "base_se_norm": [-0.2967904806137085, 0.009736776351928711], "gt_se_raw": [38.0, 48.0], "gt_se_norm": [-0.3148214817047119, -0.13451135158538818], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003775.pt"}
{"idx": 3777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8bRkdkfWv3g.mp4", "duration": 110.92, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the fruits are displayed in a supermarket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the fruits are displayed in a supermarket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 84.0], "base_se_norm": [0.06382977962493896, 0.5146051645278931], "gt_se_raw": [58.0, 77.0], "gt_se_norm": [0.04579877853393555, 0.3883880376815796], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003777.pt"}
{"idx": 3779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8bRkdkfWv3g.mp4", "duration": 110.92, "prompt_text": "Please find the visual event described by the sentence 'Show me when a boy falls off a moving truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a boy falls off a moving truck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.4230075478553772, -0.36891454458236694], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.38694554567337036, -0.35088348388671875], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003779.pt"}
{"idx": 3781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ansWZq7yULE.mp4", "duration": 63.6636, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the big turtle gives the little turtle a high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the big turtle gives the little turtle a high-five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.6230185031890869, -0.4973580241203308], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.5287731289863586, -0.43452775478363037], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003781.pt"}
{"idx": 3783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ansWZq7yULE.mp4", "duration": 63.6636, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the big turtle talks about how his kids are born and find their way to the ocean.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the big turtle talks about how his kids are born and find their way to the ocean.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 27.0], "base_se_norm": [-0.6544336080551147, -0.15179163217544556], "gt_se_raw": [38.0, 46.0], "gt_se_norm": [0.19377481937408447, 0.4450957775115967], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003783.pt"}
{"idx": 3785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HZ21eT9lyog.mp4", "duration": 31.36466666666667, "prompt_text": "Please find the visual event described by the sentence 'When is a woman using a white hand mixer in a yellow bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman using a white hand mixer in a yellow bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.4898717999458313, -0.29857373237609863], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.4898717999458313, -0.3623397946357727], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003785.pt"}
{"idx": 3787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HZ21eT9lyog.mp4", "duration": 31.36466666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman in a leopard print shirt gesturing while explaining a cooking step?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman in a leopard print shirt gesturing while explaining a cooking step?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.2753204107284546, 0.46661853790283203], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [0.2753204107284546, 0.46661853790283203], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003787.pt"}
{"idx": 3789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HZ21eT9lyog.mp4", "duration": 31.36466666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman is chopping ingredients on a green cutting board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman is chopping ingredients on a green cutting board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 19.0], "base_se_norm": [-0.04350966215133667, 0.2115544080734253], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.04350966215133667, 0.147788405418396], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003789.pt"}
{"idx": 3791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FwqsdGLhdgA.mp4", "duration": 65.44037499999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman picks up the paper from the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman picks up the paper from the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.4498808979988098, -0.38875657320022583], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.4498808979988098, -0.38875657320022583], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003791.pt"}
{"idx": 3793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FwqsdGLhdgA.mp4", "duration": 65.44037499999999, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is vacuuming aggressively out of anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is vacuuming aggressively out of anger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 46.0], "base_se_norm": [0.008551716804504395, 0.4058598279953003], "gt_se_raw": [33.0, 43.0], "gt_se_norm": [0.008551716804504395, 0.3141733407974243], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003793.pt"}
{"idx": 3795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQm5doF_Uc.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where fireworks are being set off over the castle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where fireworks are being set off over the castle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.6588910222053528, -0.6326519250869751], "gt_se_raw": [26.0, 27.0], "gt_se_norm": [-0.6588910222053528, -0.6457714438438416], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003795.pt"}
{"idx": 3797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQm5doF_Uc.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a man falls off a reindeer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a man falls off a reindeer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [-0.48833656311035156, -0.4620974063873291], "gt_se_raw": [40.0, 41.0], "gt_se_norm": [-0.47521698474884033, -0.4620974063873291], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003797.pt"}
{"idx": 3799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TbQm5doF_Uc.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the woman in the green dress and the man in the white uniform are talking in a boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the woman in the green dress and the man in the white uniform are talking in a boat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 54.0], "base_se_norm": [-0.3833799362182617, -0.2915428876876831], "gt_se_raw": [46.0, 53.0], "gt_se_norm": [-0.39649951457977295, -0.30466246604919434], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003799.pt"}
{"idx": 3801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r5L7Iokg5RY.mp4", "duration": 25.0, "prompt_text": "Please find the visual event described by the sentence 'When does the Tyrannosaurus Rex let out a roar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Tyrannosaurus Rex let out a roar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.2799999713897705, -0.12000000476837158], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.2799999713897705, -0.12000000476837158], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003801.pt"}
{"idx": 3803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r5L7Iokg5RY.mp4", "duration": 25.0, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a dinosaur is lying on the ground, defeated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a dinosaur is lying on the ground, defeated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [0.36000001430511475, 0.6000000238418579], "gt_se_raw": [16.0, 20.0], "gt_se_norm": [0.2799999713897705, 0.6000000238418579], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003803.pt"}
{"idx": 3805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZP7sEgR4Jic.mp4", "duration": 199.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the player obtain the oak sapling by breaking tree leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player obtain the oak sapling by breaking tree leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.6794657707214355, -0.6393990516662598], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.6794657707214355, -0.6393990516662598], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003805.pt"}
{"idx": 3807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZP7sEgR4Jic.mp4", "duration": 199.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the player is trying to retrieve the bones that fell off the platform.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the player is trying to retrieve the bones that fell off the platform.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 138.0], "base_se_norm": [0.2520867586135864, 0.3823038339614868], "gt_se_raw": [125.0, 132.0], "gt_se_norm": [0.2520867586135864, 0.3222036361694336], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003807.pt"}
{"idx": 3809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZP7sEgR4Jic.mp4", "duration": 199.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the tree finally growing after the player wins the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the tree finally growing after the player wins the game?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 188.0], "base_se_norm": [0.85308837890625, 0.8831385374069214], "gt_se_raw": [185.0, 186.0], "gt_se_norm": [0.85308837890625, 0.8631051778793335], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003809.pt"}
{"idx": 3811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9qsy9kvwPk.mp4", "duration": 52.986266666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the teacher places the fuzzy toy on the boy's stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the teacher places the fuzzy toy on the boy's stomach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.1695961356163025, 0.019132018089294434], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [-0.1695961356163025, 0.019132018089294434], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003811.pt"}
{"idx": 3813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s9qsy9kvwPk.mp4", "duration": 52.986266666666666, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a blue shirt touch the boy's neck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a blue shirt touch the boy's neck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 43.0], "base_se_norm": [0.3588426113128662, 0.623062014579773], "gt_se_raw": [40.0, 44.0], "gt_se_norm": [0.5098252296447754, 0.660807728767395], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003813.pt"}
{"idx": 3815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RN2P75hH74.mp4", "duration": 62.97958333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the white shirt put his hands on his face in distress?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the white shirt put his hands on his face in distress?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 41.0], "base_se_norm": [0.2384965419769287, 0.3020092248916626], "gt_se_raw": [39.0, 41.0], "gt_se_norm": [0.2384965419769287, 0.3020092248916626], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003815.pt"}
{"idx": 3817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RN2P75hH74.mp4", "duration": 62.97958333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the blue shirt eats the food with a shocked expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the blue shirt eats the food with a shocked expression.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [0.11147129535675049, 0.20674026012420654], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [0.11147129535675049, 0.20674026012420654], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003817.pt"}
{"idx": 3819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qX9FSZJu448.mp4", "duration": 294.12716666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man doing push-ups on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man doing push-ups on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 198.0], "base_se_norm": [0.3055577278137207, 0.3463563919067383], "gt_se_raw": [170.0, 174.0], "gt_se_norm": [0.15596258640289307, 0.18316173553466797], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003819.pt"}
{"idx": 3821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qX9FSZJu448.mp4", "duration": 294.12716666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man successfully run outside in a park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 232.0 - 245.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man successfully run outside in a park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 232.0 - 245.0 seconds.<|im_end|>\n", "base_se_raw": [232.0, 245.0], "base_se_norm": [0.5775489807128906, 0.6659461259841919], "gt_se_raw": [232.0, 245.0], "gt_se_norm": [0.5775489807128906, 0.6659461259841919], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003821.pt"}
{"idx": 3823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qX9FSZJu448.mp4", "duration": 294.12716666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the man is holding a cat while balancing on one leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 228.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the man is holding a cat while balancing on one leg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 228.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 228.0], "base_se_norm": [0.5163508653640747, 0.5503498315811157], "gt_se_raw": [223.0, 228.0], "gt_se_norm": [0.5163508653640747, 0.5503498315811157], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003823.pt"}
{"idx": 3825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GR2o6k8aPlI.mp4", "duration": 189.75, "prompt_text": "Please find the visual event described by the sentence 'When does the papa shark appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the papa shark appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.7364953756332397, -0.6837944984436035], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.5889328122138977, -0.5573122501373291], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003825.pt"}
{"idx": 3827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GR2o6k8aPlI.mp4", "duration": 189.75, "prompt_text": "Please find the visual event described by the sentence 'When do the little fish get scared and run away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the little fish get scared and run away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 143.0], "base_se_norm": [0.22266137599945068, 0.5072463750839233], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.23320162296295166, 0.2648221254348755], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003827.pt"}
{"idx": 3829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IDIQKPApIxM.mp4", "duration": 55.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the adult guide the girl's hand to feel the blue scribbles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the adult guide the girl's hand to feel the blue scribbles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.8909751772880554, -0.6365839242935181], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.5639006495475769, -0.4912174344062805], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003829.pt"}
{"idx": 3831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IDIQKPApIxM.mp4", "duration": 55.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the page is turned to reveal a drawing with red scribbles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the page is turned to reveal a drawing with red scribbles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.20048457384109497, -0.01877647638320923], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.1278012990951538, -0.01877647638320923], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003831.pt"}
{"idx": 3833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IDIQKPApIxM.mp4", "duration": 55.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the girl puts the end of the cane-like tool in her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the girl puts the end of the cane-like tool in her mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.7443974018096924, 0.8897638320922852], "gt_se_raw": [42.0, 44.0], "gt_se_norm": [0.5263476371765137, 0.5990308523178101], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003833.pt"}
{"idx": 3835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0hhbMi5xFtc.mp4", "duration": 163.83033333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the water start pouring out of the bottle after the cap is removed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the water start pouring out of the bottle after the cap is removed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 62.0], "base_se_norm": [-0.32857370376586914, -0.24311941862106323], "gt_se_raw": [54.0, 62.0], "gt_se_norm": [-0.34078145027160645, -0.24311941862106323], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003835.pt"}
{"idx": 3837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0hhbMi5xFtc.mp4", "duration": 163.83033333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man pierces the soda can, causing it to spray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man pierces the soda can, causing it to spray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 112.0], "base_se_norm": [0.2940216064453125, 0.36726808547973633], "gt_se_raw": [110.0, 114.0], "gt_se_norm": [0.3428525924682617, 0.39168357849121094], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003837.pt"}
{"idx": 3839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkQyRE0byBI.mp4", "duration": 155.86404166666665, "prompt_text": "Please find the visual event described by the sentence 'When does the man enter the house and start exploring the dark room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man enter the house and start exploring the dark room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 45.0], "base_se_norm": [-0.9871683120727539, -0.42257368564605713], "gt_se_raw": [24.0, 43.0], "gt_se_norm": [-0.6920393109321594, -0.4482370615005493], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003839.pt"}
{"idx": 3841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkQyRE0byBI.mp4", "duration": 155.86404166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man punch the window, causing the glass to shatter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man punch the window, causing the glass to shatter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.5893857479095459, -0.5508906841278076], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.24292993545532227, -0.20443487167358398], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003841.pt"}
{"idx": 3843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkQyRE0byBI.mp4", "duration": 155.86404166666665, "prompt_text": "Please find the visual event described by the sentence 'When do the man and woman have a verbal confrontation after he kisses her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the man and woman have a verbal confrontation after he kisses her?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 143.0], "base_se_norm": [0.15485262870788574, 0.8349324464797974], "gt_se_raw": [87.0, 111.0], "gt_se_norm": [0.11635756492614746, 0.42431819438934326], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003843.pt"}
{"idx": 3845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-8qvGAWiPQY.mp4", "duration": 35.32695833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a man in a suit is speaking in a hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a man in a suit is speaking in a hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 17.0], "base_se_norm": [-0.547088086605072, -0.03756219148635864], "gt_se_raw": [8.0, 17.0], "gt_se_norm": [-0.547088086605072, -0.03756219148635864], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003845.pt"}
{"idx": 3847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-8qvGAWiPQY.mp4", "duration": 35.32695833333333, "prompt_text": "Please find the visual event described by the sentence 'Where is the moment when a man is talking in front of a group of people on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where is the moment when a man is talking in front of a group of people on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [0.35873568058013916, 0.6984196901321411], "gt_se_raw": [23.0, 30.0], "gt_se_norm": [0.30212175846099854, 0.6984196901321411], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003847.pt"}
{"idx": 3849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RxAtuMu_ph4.mp4", "duration": 128.6285, "prompt_text": "Please find the visual event described by the sentence 'When can we see two women pointing guns at each other in a room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two women pointing guns at each other in a room?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 55.0], "base_se_norm": [-0.1914699673652649, -0.14482402801513672], "gt_se_raw": [50.0, 56.0], "gt_se_norm": [-0.22256726026535034, -0.129275381565094], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003849.pt"}
{"idx": 3851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RxAtuMu_ph4.mp4", "duration": 128.6285, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when two women are fighting in a kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when two women are fighting in a kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 64.0], "base_se_norm": [-0.16037267446517944, -0.0048860907554626465], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.1137266755104065, -0.020434796810150146], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003851.pt"}
{"idx": 3853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RxAtuMu_ph4.mp4", "duration": 128.6285, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of a woman in a white suit falling through the sky amidst debris.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of a woman in a white suit falling through the sky amidst debris.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 116.0], "base_se_norm": [0.7103519439697266, 0.8036439418792725], "gt_se_raw": [111.0, 117.0], "gt_se_norm": [0.7259006500244141, 0.8191925287246704], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003853.pt"}
{"idx": 3855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-UggNH1YeRY.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in green put two cans of yams into her cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in green put two cans of yams into her cart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 77.0], "base_se_norm": [-0.2417982816696167, -0.18914532661437988], "gt_se_raw": [73.0, 76.0], "gt_se_norm": [-0.23126769065856934, -0.19967591762542725], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003855.pt"}
{"idx": 3857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-UggNH1YeRY.mp4", "duration": 189.92306666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman in red rams her cart into the other woman's cart, knocking it over.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman in red rams her cart into the other woman's cart, knocking it over.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 141.0], "base_se_norm": [0.4426894187927246, 0.48481178283691406], "gt_se_raw": [137.0, 139.0], "gt_se_norm": [0.4426894187927246, 0.46375060081481934], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003857.pt"}
{"idx": 3859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6CfZwFzapCo.mp4", "duration": 113.19641666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the customer rudely yells to get the barista's attention.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the customer rudely yells to get the barista's attention.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 59.0], "base_se_norm": [-0.11657983064651489, 0.04243576526641846], "gt_se_raw": [55.0, 59.0], "gt_se_norm": [-0.02823781967163086, 0.04243576526641846], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003859.pt"}
{"idx": 3861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6CfZwFzapCo.mp4", "duration": 113.19641666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man in the cap receives his coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man in the cap receives his coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.7349739074707031, -0.6819687485694885], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.7173055410385132, -0.6819687485694885], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003861.pt"}
{"idx": 3863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6CfZwFzapCo.mp4", "duration": 113.19641666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the two baristas laugh after playing a prank on the rude customer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the two baristas laugh after playing a prank on the rude customer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 72.0], "base_se_norm": [0.21911978721618652, 0.2721250057220459], "gt_se_raw": [70.0, 72.0], "gt_se_norm": [0.23678827285766602, 0.2721250057220459], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003863.pt"}
{"idx": 3865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BlJaw7xc1bw.mp4", "duration": 190.2902902902903, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the banana is mashed with a wooden tool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the banana is mashed with a wooden tool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 47.0], "base_se_norm": [-0.7267333269119263, -0.5060178637504578], "gt_se_raw": [27.0, 48.0], "gt_se_norm": [-0.7162230014801025, -0.4955075979232788], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003865.pt"}
{"idx": 3867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BlJaw7xc1bw.mp4", "duration": 190.2902902902903, "prompt_text": "Please find the visual event described by the sentence 'At what point is the banana cut into small pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the banana cut into small pieces?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 138.0], "base_se_norm": [0.11408734321594238, 0.45041561126708984], "gt_se_raw": [106.0, 138.0], "gt_se_norm": [0.11408734321594238, 0.45041561126708984], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003867.pt"}
{"idx": 3869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XwKm4RILZa4.mp4", "duration": 60.8608, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video demonstrate the process of searching for a YouTube video to describe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video demonstrate the process of searching for a YouTube video to describe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.375624418258667, -0.3099006414413452], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.375624418258667, -0.3099006414413452], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003869.pt"}
{"idx": 3871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XwKm4RILZa4.mp4", "duration": 60.8608, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the video explains that described content can be embedded on your own website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the video explains that described content can be embedded on your own website.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [0.051579952239990234, 0.1830275058746338], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [0.051579952239990234, 0.15016555786132812], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003871.pt"}
{"idx": 3873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XwKm4RILZa4.mp4", "duration": 60.8608, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the interface for recording an audio description?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the interface for recording an audio description?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.6713812351226807, -0.507071852684021], "gt_se_raw": [21.0, 25.0], "gt_se_norm": [-0.3099006414413452, -0.17845314741134644], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003873.pt"}
{"idx": 3875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g4woIrtHTOc.mp4", "duration": 149.16, "prompt_text": "Please find the visual event described by the sentence 'Show the view of the Shanghai University campus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the view of the Shanghai University campus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.7586484551429749, -0.6513810753822327], "gt_se_raw": [19.0, 25.0], "gt_se_norm": [-0.7452399730682373, -0.6647894978523254], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003875.pt"}
{"idx": 3877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g4woIrtHTOc.mp4", "duration": 149.16, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the lecture theatres in The University of Sydney are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the lecture theatres in The University of Sydney are shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 96.0], "base_se_norm": [-0.2223116159439087, 0.28720831871032715], "gt_se_raw": [61.0, 78.0], "gt_se_norm": [-0.18208634853363037, 0.04585671424865723], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003877.pt"}
{"idx": 3879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QoWskNBiHdE.mp4", "duration": 152.31883333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the three women chatting at the picnic table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the three women chatting at the picnic table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 50.0], "base_se_norm": [-0.6980018615722656, -0.3434823751449585], "gt_se_raw": [22.0, 51.0], "gt_se_norm": [-0.7111322283744812, -0.3303520083427429], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003879.pt"}
{"idx": 3881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QoWskNBiHdE.mp4", "duration": 152.31883333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the 15% statistic from the United Nations is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the 15% statistic from the United Nations is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 63.0], "base_se_norm": [-0.3303520083427429, -0.17278778553009033], "gt_se_raw": [54.0, 63.0], "gt_se_norm": [-0.29096096754074097, -0.17278778553009033], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003881.pt"}
{"idx": 3883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QoWskNBiHdE.mp4", "duration": 152.31883333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man introduces himself and the video's topic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man introduces himself and the video's topic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 22.0], "base_se_norm": [-0.9737392663955688, -0.7111322283744812], "gt_se_raw": [3.0, 22.0], "gt_se_norm": [-0.960608959197998, -0.7111322283744812], "start_token_span": [75, 78], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003883.pt"}
{"idx": 3885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLackFf0Zjw.mp4", "duration": 79.78803333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the knight in black armor decapitates the horse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the knight in black armor decapitates the horse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 27.0], "base_se_norm": [-0.37333959341049194, -0.3232067823410034], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.37333959341049194, -0.3232067823410034], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003885.pt"}
{"idx": 3887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLackFf0Zjw.mp4", "duration": 79.78803333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the knight with the purple lightsaber fight the knight with the red lightsaber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the knight with the purple lightsaber fight the knight with the red lightsaber?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 66.0], "base_se_norm": [0.10292232036590576, 0.6543834209442139], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [0.2533208131790161, 0.3285200595855713], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003887.pt"}
{"idx": 3889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpeXE9XCU3o.mp4", "duration": 54.541666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me when the word 'creative' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the word 'creative' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.889992356300354], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.8533231616020203], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003889.pt"}
{"idx": 3891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpeXE9XCU3o.mp4", "duration": 54.541666666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see the columns in front of Truman Hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the columns in front of Truman Hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 25.0], "base_se_norm": [-0.26661574840545654, -0.08326971530914307], "gt_se_raw": [20.0, 25.0], "gt_se_norm": [-0.26661574840545654, -0.08326971530914307], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003891.pt"}
{"idx": 3893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hpeXE9XCU3o.mp4", "duration": 54.541666666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the self-portrait of a man's shadow holding a white cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the self-portrait of a man's shadow holding a white cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [0.3567608594894409, 0.5401068925857544], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [0.39343011379241943, 0.5767761468887329], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003893.pt"}
{"idx": 3895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZYYCSEV-i1Y.mp4", "duration": 151.15099999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the woman in the purple dress and green scarf is talking in a meeting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the woman in the purple dress and green scarf is talking in a meeting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 69.0], "base_se_norm": [-0.19286012649536133, -0.08700573444366455], "gt_se_raw": [60.0, 63.0], "gt_se_norm": [-0.20609194040298462, -0.16639649868011475], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003895.pt"}
{"idx": 3897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZYYCSEV-i1Y.mp4", "duration": 151.15099999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the woman fall onto the round ottoman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman fall onto the round ottoman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 102.0], "base_se_norm": [0.32318007946014404, 0.3496437072753906], "gt_se_raw": [99.0, 101.0], "gt_se_norm": [0.3099483251571655, 0.3364119529724121], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003897.pt"}
{"idx": 3899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/noGcxr9xiGE.mp4", "duration": 56.99026666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the boy's shoes are untied.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the boy's shoes are untied.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.7192503213882446, -0.5788754224777222], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.7192503213882446, -0.5788754224777222], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003899.pt"}
{"idx": 3901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/noGcxr9xiGE.mp4", "duration": 56.99026666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boy swing one lace around the other to form the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy swing one lace around the other to form the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 35.0], "base_se_norm": [-0.017376065254211426, 0.2282799482345581], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [0.08790504932403564, 0.2282799482345581], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003901.pt"}
{"idx": 3903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/noGcxr9xiGE.mp4", "duration": 56.99026666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the boy smiles, looking satisfied with his work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the boy smiles, looking satisfied with his work.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [0.36865484714508057, 0.47393596172332764], "gt_se_raw": [47.0, 51.0], "gt_se_norm": [0.6494045257568359, 0.7897793054580688], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003903.pt"}
{"idx": 3905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZAeKfGpd2M.mp4", "duration": 230.60537499999998, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the exterior of the 'Republic of Booza' shop is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the exterior of the 'Republic of Booza' shop is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.8352163434028625, -0.8091978430747986], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.8438891768455505, -0.8178706765174866], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003905.pt"}
{"idx": 3907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oZAeKfGpd2M.mp4", "duration": 230.60537499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point do two women try the white-colored 'Qashta' ice cream together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two women try the white-colored 'Qashta' ice cream together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 149.0], "base_se_norm": [0.23154115676879883, 0.2922508716583252], "gt_se_raw": [147.0, 149.0], "gt_se_norm": [0.2749052047729492, 0.2922508716583252], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003907.pt"}
{"idx": 3909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4swYF9Szvs.mp4", "duration": 173.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the word 'qu\u00e9' and its English translation 'what' displayed on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the word 'qu\u00e9' and its English translation 'what' displayed on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 13.0], "base_se_norm": [-0.9194630980491638, -0.8504314422607422], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.8964525461196899, -0.8504314422607422], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003909.pt"}
{"idx": 3911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4swYF9Szvs.mp4", "duration": 173.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman appear celebrating her birthday with a cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman appear celebrating her birthday with a cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 59.0], "base_se_norm": [-0.4247363209724426, -0.32118886709213257], "gt_se_raw": [50.0, 60.0], "gt_se_norm": [-0.4247363209724426, -0.30968356132507324], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003911.pt"}
{"idx": 3913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4swYF9Szvs.mp4", "duration": 173.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the word 'C\u00f3mo' and its translation are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the word 'C\u00f3mo' and its translation are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 103.0], "base_se_norm": [0.12751686573028564, 0.18504321575164795], "gt_se_raw": [99.0, 103.0], "gt_se_norm": [0.1390221118927002, 0.18504321575164795], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003913.pt"}
{"idx": 3915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jF3I7VpfCEU.mp4", "duration": 174.96645555555554, "prompt_text": "Please find the visual event described by the sentence 'Show me when the men display the words 'Fuck Face' and 'Clusterfuck'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the men display the words 'Fuck Face' and 'Clusterfuck'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 60.0], "base_se_norm": [-0.4284619092941284, -0.3141542673110962], "gt_se_raw": [51.0, 57.0], "gt_se_norm": [-0.41703110933303833, -0.3484465479850769], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003915.pt"}
{"idx": 3917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jF3I7VpfCEU.mp4", "duration": 174.96645555555554, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the on-screen question 'What was Thomas' high school superlative?' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the on-screen question 'What was Thomas' high school superlative?' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 105.0], "base_se_norm": [0.15450692176818848, 0.20023000240325928], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [0.15450692176818848, 0.20023000240325928], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003917.pt"}
{"idx": 3919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qI34nBlJxP8.mp4", "duration": 104.06229166666665, "prompt_text": "Please find the visual event described by the sentence 'When does Snoopy put on his pilot goggles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Snoopy put on his pilot goggles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.38498377799987793, -0.3081067204475403], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.38498377799987793, -0.3273259997367859], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003919.pt"}
{"idx": 3921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qI34nBlJxP8.mp4", "duration": 104.06229166666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is Snoopy's doghouse flying near the Eiffel Tower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is Snoopy's doghouse flying near the Eiffel Tower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 58.0], "base_se_norm": [0.018620610237121582, 0.11471688747406006], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.03783988952636719, 0.0762784481048584], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003921.pt"}
{"idx": 3923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qI34nBlJxP8.mp4", "duration": 104.06229166666665, "prompt_text": "Please find the visual event described by the sentence 'When does Charlie Brown have a popcorn bucket stuck on his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Charlie Brown have a popcorn bucket stuck on his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 97.0], "base_se_norm": [0.7489523887634277, 0.8642679452896118], "gt_se_raw": [92.0, 95.0], "gt_se_norm": [0.7681716680526733, 0.8258293867111206], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003923.pt"}
{"idx": 3925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y4yZelVCE3o.mp4", "duration": 180.52, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the family runs out of the destroyed house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the family runs out of the destroyed house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 42.0], "base_se_norm": [-0.601152241230011, -0.5346776247024536], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.601152241230011, -0.5679149627685547], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003925.pt"}
{"idx": 3927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y4yZelVCE3o.mp4", "duration": 180.52, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a giant donut sign rolls down the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a giant donut sign rolls down the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 118.0], "base_se_norm": [0.2740969657897949, 0.307334303855896], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [0.2740969657897949, 0.307334303855896], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003927.pt"}
{"idx": 3929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D0a0aNqTehM.mp4", "duration": 360.72703333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the caterpillar dancing with the cocoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the caterpillar dancing with the cocoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 62.0], "base_se_norm": [-0.8336692452430725, -0.6562497615814209], "gt_se_raw": [47.0, 51.0], "gt_se_norm": [-0.739415168762207, -0.7172377109527588], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003929.pt"}
{"idx": 3931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D0a0aNqTehM.mp4", "duration": 360.72703333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the green caterpillar being wrapped in a cocoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the green caterpillar being wrapped in a cocoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 85.0], "base_se_norm": [-0.6174392700195312, -0.5287294983863831], "gt_se_raw": [68.0, 85.0], "gt_se_norm": [-0.6229835748672485, -0.5287294983863831], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003931.pt"}
{"idx": 3933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D0a0aNqTehM.mp4", "duration": 360.72703333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the caterpillar becomes a butterfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 254.0 - 293.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the caterpillar becomes a butterfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 254.0 - 293.0 seconds.<|im_end|>\n", "base_se_raw": [254.0, 293.0], "base_se_norm": [0.40826714038848877, 0.6244970560073853], "gt_se_raw": [292.0, 296.0], "gt_se_norm": [0.618952751159668, 0.6411302089691162], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003933.pt"}
{"idx": 3935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pjJEXkbeL-o.mp4", "duration": 278.76, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the uniform walk into the restroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the uniform walk into the restroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [-0.5336490273475647, -0.48342663049697876], "gt_se_raw": [70.0, 74.0], "gt_se_norm": [-0.4977758526802063, -0.46907734870910645], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003935.pt"}
{"idx": 3937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pjJEXkbeL-o.mp4", "duration": 278.76, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man in the tuxedo is being handcuffed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 220.0 - 236.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man in the tuxedo is being handcuffed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 220.0 - 236.0 seconds.<|im_end|>\n", "base_se_raw": [220.0, 236.0], "base_se_norm": [0.5784186124801636, 0.6932127475738525], "gt_se_raw": [236.0, 238.0], "gt_se_norm": [0.6932127475738525, 0.7075619697570801], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003937.pt"}
{"idx": 3939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fNINO2RnYz0.mp4", "duration": 14.014000000000001, "prompt_text": "Please find the visual event described by the sentence 'When can we see the colorful lion logo on a white background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the colorful lion logo on a white background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.0009989738464355469], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.0009989738464355469], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003939.pt"}
{"idx": 3941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fNINO2RnYz0.mp4", "duration": 14.014000000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the interval where the colorful text 'TOKYO MX' is forming at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the interval where the colorful text 'TOKYO MX' is forming at the bottom of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.8572855591773987, -0.4291422963142395], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.7145711183547974], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003941.pt"}
{"idx": 3943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fNINO2RnYz0.mp4", "duration": 14.014000000000001, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the loading text changes to 'Loading complete'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the loading text changes to 'Loading complete'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 13.0], "base_se_norm": [0.7125731706619263, 0.8552875518798828], "gt_se_raw": [13.0, 14.0], "gt_se_norm": [0.8552875518798828, 0.9980020523071289], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003943.pt"}
{"idx": 3945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/USGLDwaLnaI.mp4", "duration": 88.75533333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a young girl holds up a sign that reads 'Culture is'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a young girl holds up a sign that reads 'Culture is'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.3239842653274536, -0.25638270378112793], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.3239842653274536, -0.25638270378112793], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003945.pt"}
{"idx": 3947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/USGLDwaLnaI.mp4", "duration": 88.75533333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man is holding up a dark blue T-shirt with the words 'North Lawndale Pride'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man is holding up a dark blue T-shirt with the words 'North Lawndale Pride'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [0.10415899753570557, 0.19429445266723633], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [0.10415899753570557, 0.17176055908203125], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003947.pt"}
{"idx": 3949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GYDaVvjI5kY.mp4", "duration": 202.06853333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the girl catch a lizard from the rock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl catch a lizard from the rock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 68.0], "base_se_norm": [-0.3665515184402466, -0.3269609808921814], "gt_se_raw": [65.0, 67.0], "gt_se_norm": [-0.3566538691520691, -0.3368586301803589], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003949.pt"}
{"idx": 3951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GYDaVvjI5kY.mp4", "duration": 202.06853333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the two girls are standing face-to-face, talking to each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the two girls are standing face-to-face, talking to each other.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 127.0], "base_se_norm": [0.21740877628326416, 0.25699925422668457], "gt_se_raw": [124.0, 126.0], "gt_se_norm": [0.22730636596679688, 0.24710166454315186], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003951.pt"}
{"idx": 3953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GYDaVvjI5kY.mp4", "duration": 202.06853333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman discover the lizard in her hair and start screaming frantically?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 177.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman discover the lizard in her hair and start screaming frantically?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 177.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 177.0], "base_se_norm": [0.6331093311309814, 0.7518808841705322], "gt_se_raw": [169.0, 177.0], "gt_se_norm": [0.6726999282836914, 0.7518808841705322], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003953.pt"}
{"idx": 3955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ur83i6_BjbE.mp4", "duration": 143.26812222222222, "prompt_text": "Please find the visual event described by the sentence 'At what point do the man and woman slow dance together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the man and woman slow dance together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.3438875675201416, -0.27408838272094727], "gt_se_raw": [52.0, 53.0], "gt_se_norm": [-0.27408838272094727, -0.26012855768203735], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003955.pt"}
{"idx": 3957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ur83i6_BjbE.mp4", "duration": 143.26812222222222, "prompt_text": "Please find the visual event described by the sentence 'When is the woman with her hair in a bun lifting weights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman with her hair in a bun lifting weights?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 88.0], "base_se_norm": [0.20054614543914795, 0.22846579551696777], "gt_se_raw": [85.0, 87.0], "gt_se_norm": [0.18658626079559326, 0.21450603008270264], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003957.pt"}
{"idx": 3959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/83Wu5xmstn4.mp4", "duration": 132.88275, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a man teaching with diagrams on a whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a man teaching with diagrams on a whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.6989827156066895, -0.6538301706314087], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.6989827156066895, -0.6538301706314087], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003959.pt"}
{"idx": 3961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/83Wu5xmstn4.mp4", "duration": 132.88275, "prompt_text": "Please find the visual event described by the sentence 'Show the graduates sitting in the auditorium before the ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the graduates sitting in the auditorium before the ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.9096948504447937, -0.8645422458648682], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.8946439623832703, -0.8645422458648682], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003961.pt"}
{"idx": 3963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/83Wu5xmstn4.mp4", "duration": 132.88275, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the woman with short grey hair and glasses speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the woman with short grey hair and glasses speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 77.0], "base_se_norm": [0.128814697265625, 0.15891635417938232], "gt_se_raw": [74.0, 77.0], "gt_se_norm": [0.11376380920410156, 0.15891635417938232], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003963.pt"}
{"idx": 3965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xoPJu25ZHxM.mp4", "duration": 273.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the man in the red shirt flexing his bicep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man in the red shirt flexing his bicep.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 39.0], "base_se_norm": [-0.7662811875343323, -0.7151552438735962], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.5398660898208618, -0.5252586603164673], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003965.pt"}
{"idx": 3967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xoPJu25ZHxM.mp4", "duration": 273.8333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman with long dark hair, holding a glowing shell necklace, appears in a doorway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 207.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman with long dark hair, holding a glowing shell necklace, appears in a doorway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 207.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 207.0], "base_se_norm": [0.35118675231933594, 0.5118684768676758], "gt_se_raw": [185.0, 191.0], "gt_se_norm": [0.35118675231933594, 0.39500904083251953], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003967.pt"}
{"idx": 3969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBiE0ltuFko.mp4", "duration": 130.1, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl sliding down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl sliding down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 51.0], "base_se_norm": [-0.6156802773475647, -0.21598774194717407], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.21598774194717407, -0.16986936330795288], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003969.pt"}
{"idx": 3971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBiE0ltuFko.mp4", "duration": 130.1, "prompt_text": "Please find the visual event described by the sentence 'Show the girl climbing up the slide for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the girl climbing up the slide for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 129.0], "base_se_norm": [0.7986164093017578, 0.983089804649353], "gt_se_raw": [118.0, 127.0], "gt_se_norm": [0.8139891624450684, 0.9523442983627319], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003971.pt"}
{"idx": 3973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CBiE0ltuFko.mp4", "duration": 130.1, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the girl is positioning herself to get ready to slide down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the girl is positioning herself to get ready to slide down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 38.0], "base_se_norm": [-0.6617985963821411, -0.4158340096473694], "gt_se_raw": [24.0, 38.0], "gt_se_norm": [-0.63105309009552, -0.4158340096473694], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003973.pt"}
{"idx": 3975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uQp0Eihw2WA.mp4", "duration": 60.026633333333336, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the girl points her finger up to the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the girl points her finger up to the ceiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [0.8325198888778687, 0.9324754476547241], "gt_se_raw": [56.0, 58.0], "gt_se_norm": [0.8658384084701538, 0.9324754476547241], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003975.pt"}
{"idx": 3977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uQp0Eihw2WA.mp4", "duration": 60.026633333333336, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the girl waving her hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the girl waving her hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.3669477105140686, -0.2669920325279236], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.3669477105140686, -0.3003106117248535], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003977.pt"}
{"idx": 3979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0Ue-JMqh0CE.mp4", "duration": 17.684333333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see two people playing the electronic drums together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two people playing the electronic drums together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, 0.8095111846923828], "gt_se_raw": [0.0, 15.0], "gt_se_norm": [-1.0, 0.6964167356491089], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003979.pt"}
{"idx": 3981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0Ue-JMqh0CE.mp4", "duration": 17.684333333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when the drummer in the blue sweatshirt is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when the drummer in the blue sweatshirt is laughing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 15.0], "base_se_norm": [-0.0952444076538086, 0.6964167356491089], "gt_se_raw": [10.0, 15.0], "gt_se_norm": [0.13094449043273926, 0.6964167356491089], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003981.pt"}
{"idx": 3983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0Ue-JMqh0CE.mp4", "duration": 17.684333333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a woman in a green t-shirt approaches the drummer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a woman in a green t-shirt approaches the drummer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [0.583322286605835, 0.8095111846923828], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [0.6964167356491089, 0.8095111846923828], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003983.pt"}
{"idx": 3985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zkCoaPOsw5U.mp4", "duration": 197.0, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman is putting on her yellow running shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman is putting on her yellow running shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 68.0], "base_se_norm": [-0.4111675024032593, -0.3096446990966797], "gt_se_raw": [59.0, 67.0], "gt_se_norm": [-0.4010152220726013, -0.31979697942733765], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003985.pt"}
{"idx": 3987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zkCoaPOsw5U.mp4", "duration": 197.0, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the woman dives into the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the woman dives into the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 117.0], "base_se_norm": [0.15736043453216553, 0.18781721591949463], "gt_se_raw": [114.0, 116.0], "gt_se_norm": [0.15736043453216553, 0.17766499519348145], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003987.pt"}
{"idx": 3989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VIglbPLTzVs.mp4", "duration": 23.289933333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the person reading from a page that contains only Braille text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person reading from a page that contains only Braille text?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.4847559332847595], "gt_se_raw": [0.0, 6.0], "gt_se_norm": [-1.0, -0.4847559332847595], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003989.pt"}
{"idx": 3991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VIglbPLTzVs.mp4", "duration": 23.289933333333334, "prompt_text": "Please find the visual event described by the sentence 'At what time is the page with the tactile illustration of a person visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the page with the tactile illustration of a person visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 23.0], "base_se_norm": [-0.3988819122314453, 0.9751021862030029], "gt_se_raw": [8.0, 23.0], "gt_se_norm": [-0.3130079507827759, 0.9751021862030029], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003991.pt"}
{"idx": 3993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VIglbPLTzVs.mp4", "duration": 23.289933333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the person's hand is touching the tactile drawing of the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the person's hand is touching the tactile drawing of the girl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [0.7174801826477051, 0.9751021862030029], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [0.7174801826477051, 0.8892282247543335], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003993.pt"}
{"idx": 3995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JEDrAOt0er8.mp4", "duration": 94.55279166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the four-armed woman emerges from the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the four-armed woman emerges from the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 35.0], "base_se_norm": [-0.619260311126709, -0.2596728801727295], "gt_se_raw": [20.0, 34.0], "gt_se_norm": [-0.5769559144973755, -0.28082507848739624], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003995.pt"}
{"idx": 3997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JEDrAOt0er8.mp4", "duration": 94.55279166666666, "prompt_text": "Please find the visual event described by the sentence 'At what time does the woman dance to the music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the woman dance to the music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 76.0], "base_se_norm": [0.1422189474105835, 0.607567548751831], "gt_se_raw": [53.0, 66.0], "gt_se_norm": [0.12106680870056152, 0.396045446395874], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003997.pt"}
{"idx": 3999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ho8L7wzkeG0.mp4", "duration": 92.29220000000001, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman introduces herself in front of the castle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman introduces herself in front of the castle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8699781894683838, -0.761626660823822], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8483078479766846, -0.7832969427108765], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00003999.pt"}
{"idx": 4001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ho8L7wzkeG0.mp4", "duration": 92.29220000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a group of people are in a meeting looking at scientific data on laptops.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a group of people are in a meeting looking at scientific data on laptops.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 58.0], "base_se_norm": [0.08351516723632812, 0.2568776607513428], "gt_se_raw": [50.0, 58.0], "gt_se_norm": [0.08351516723632812, 0.2568776607513428], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004001.pt"}
{"idx": 4003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ho8L7wzkeG0.mp4", "duration": 92.29220000000001, "prompt_text": "Please find the visual event described by the sentence 'When is the woman standing next to a man playing the bagpipes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman standing next to a man playing the bagpipes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [-0.4799126982688904, -0.3715611696243286], "gt_se_raw": [64.0, 74.0], "gt_se_norm": [0.386899471282959, 0.6036025285720825], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004003.pt"}
{"idx": 4005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3h6KMumLAvI.mp4", "duration": 15.849166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does a person pick up a stack of cards from the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a person pick up a stack of cards from the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [0.5142751932144165, 0.8928439617156982], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [0.5142751932144165, 0.766654372215271], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004005.pt"}
{"idx": 4007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3h6KMumLAvI.mp4", "duration": 15.849166666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person on the left begins to actively guide the other person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person on the left begins to actively guide the other person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 12.0], "base_se_norm": [-0.6214312314987183, 0.5142751932144165], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.6214312314987183, -0.369051992893219], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004007.pt"}
{"idx": 4009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gsYL4PC0hyk.mp4", "duration": 253.16958333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the five stages of grief listed on the blackboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the five stages of grief listed on the blackboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 66.0], "base_se_norm": [-0.5260094404220581, -0.4786103367805481], "gt_se_raw": [57.0, 66.0], "gt_se_norm": [-0.5497089624404907, -0.4786103367805481], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004009.pt"}
{"idx": 4011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gsYL4PC0hyk.mp4", "duration": 253.16958333333332, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a character is seen riding a rollercoaster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a character is seen riding a rollercoaster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 83.0], "base_se_norm": [-0.40751177072525024, -0.34431302547454834], "gt_se_raw": [74.0, 83.0], "gt_se_norm": [-0.4154115915298462, -0.34431302547454834], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004011.pt"}
{"idx": 4013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gsYL4PC0hyk.mp4", "duration": 253.16958333333332, "prompt_text": "Please find the visual event described by the sentence 'When is the phrase 'Do things that make you happy' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the phrase 'Do things that make you happy' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 141.0], "base_se_norm": [-0.10731774568557739, 0.11387789249420166], "gt_se_raw": [113.0, 115.0], "gt_se_norm": [-0.10731774568557739, -0.09151804447174072], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004013.pt"}
{"idx": 4015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMmIMQuqVa0.mp4", "duration": 136.87006666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a girl in a purple sweater plays with a green balloon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a girl in a purple sweater plays with a green balloon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 21.0], "base_se_norm": [-0.8392636179924011, -0.6931396722793579], "gt_se_raw": [11.0, 21.0], "gt_se_norm": [-0.8392636179924011, -0.6931396722793579], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004015.pt"}
{"idx": 4017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rMmIMQuqVa0.mp4", "duration": 136.87006666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman in a blue jacket interacts with a boy outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman in a blue jacket interacts with a boy outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 83.0], "base_se_norm": [0.008255481719970703, 0.2128291130065918], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [0.008255481719970703, 0.08131742477416992], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004017.pt"}
{"idx": 4019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xdr_rWE8qOQ.mp4", "duration": 195.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the duck eating green peas from a bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the duck eating green peas from a bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.8257006406784058, -0.7949419021606445], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.8257006406784058, -0.8051947951316833], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004019.pt"}
{"idx": 4021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xdr_rWE8qOQ.mp4", "duration": 195.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a dog and ducks are gathered around a slice of watermelon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a dog and ducks are gathered around a slice of watermelon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 124.0], "base_se_norm": [0.2406015396118164, 0.27136027812957764], "gt_se_raw": [122.0, 124.0], "gt_se_norm": [0.25085437297821045, 0.27136027812957764], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004021.pt"}
{"idx": 4023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xdr_rWE8qOQ.mp4", "duration": 195.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the duck is riding on a robotic vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the duck is riding on a robotic vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 164.0], "base_se_norm": [0.6609705686569214, 0.6814764738082886], "gt_se_raw": [162.0, 165.0], "gt_se_norm": [0.6609705686569214, 0.6917293071746826], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004023.pt"}
{"idx": 4025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CMYlBq83gu4.mp4", "duration": 155.155, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the man is writing on a check while viewing it on a large screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the man is writing on a check while viewing it on a large screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 106.0], "base_se_norm": [0.18591082096099854, 0.36637556552886963], "gt_se_raw": [101.0, 106.0], "gt_se_norm": [0.30192387104034424, 0.36637556552886963], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004025.pt"}
{"idx": 4027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CMYlBq83gu4.mp4", "duration": 155.155, "prompt_text": "Please find the visual event described by the sentence 'When is the fountain in the courtyard shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the fountain in the courtyard shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 143.0], "base_se_norm": [0.7917566299438477, 0.843317985534668], "gt_se_raw": [139.0, 143.0], "gt_se_norm": [0.7917566299438477, 0.843317985534668], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004027.pt"}
{"idx": 4029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghg2AP3i5TI.mp4", "duration": 34.26756666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a red T-shirt writing at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a red T-shirt writing at the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 25.0], "base_se_norm": [0.22564876079559326, 0.45910561084747314], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [0.22564876079559326, 0.40074145793914795], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004029.pt"}
{"idx": 4031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghg2AP3i5TI.mp4", "duration": 34.26756666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time is a woman with a bright red scarf seen talking to someone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a woman with a bright red scarf seen talking to someone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.12453663349151611, 0.050556063652038574], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [-0.12453663349151611, -0.06617236137390137], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004031.pt"}
{"idx": 4033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghg2AP3i5TI.mp4", "duration": 34.26756666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where people are gathered and looking at the posters on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where people are gathered and looking at the posters on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 30.0], "base_se_norm": [-0.7665430903434753, 0.7509267330169678], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.5914504528045654, -0.41635775566101074], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004033.pt"}
{"idx": 4035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bhA2zrVBikA.mp4", "duration": 128.9, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman on the right hand a packet to the woman on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman on the right hand a packet to the woman on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [0.024049758911132812, 0.10162925720214844], "gt_se_raw": [67.0, 71.0], "gt_se_norm": [0.0395655632019043, 0.10162925720214844], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004035.pt"}
{"idx": 4037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bhA2zrVBikA.mp4", "duration": 128.9, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the woman on the left praises the other woman, saying 'Good looking'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the woman on the left praises the other woman, saying 'Good looking'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 111.0], "base_se_norm": [0.6602017879486084, 0.722265362739563], "gt_se_raw": [119.0, 121.0], "gt_se_norm": [0.8463926315307617, 0.8774244785308838], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004037.pt"}
{"idx": 4039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t-WR-8-HEx4.mp4", "duration": 31.197833333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the man walking down the hallway using his cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man walking down the hallway using his cane?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 29.0], "base_se_norm": [-1.0, 0.8591034412384033], "gt_se_raw": [1.0, 23.0], "gt_se_norm": [-0.9358929991722107, 0.47446131706237793], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004039.pt"}
{"idx": 4041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t-WR-8-HEx4.mp4", "duration": 31.197833333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the interval where the woman stands next to the man while he is leaning against the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the interval where the woman stands next to the man while he is leaning against the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.1666088104248047, 0.15392625331878662], "gt_se_raw": [12.0, 18.0], "gt_se_norm": [-0.230715811252594, 0.15392625331878662], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004041.pt"}
{"idx": 4043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t-WR-8-HEx4.mp4", "duration": 31.197833333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the man is opening the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the man is opening the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.47446131706237793, 0.7949963808059692], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [0.7308894395828247, 0.9232103824615479], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004043.pt"}
{"idx": 4045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nuavO7Of92M.mp4", "duration": 142.97616666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a person is writing on a whiteboard with a green marker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a person is writing on a whiteboard with a green marker.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 24.0], "base_se_norm": [-0.7342214584350586, -0.6642796993255615], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.7342214584350586, -0.706244707107544], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004045.pt"}
{"idx": 4047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nuavO7Of92M.mp4", "duration": 142.97616666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the woman explains that the program utilizes a professional development school model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the woman explains that the program utilizes a professional development school model.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 76.0], "base_se_norm": [-0.07676917314529419, 0.06311428546905518], "gt_se_raw": [67.0, 69.0], "gt_se_norm": [-0.06278085708618164, -0.03480416536331177], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004047.pt"}
{"idx": 4049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksdOh7MbRgk.mp4", "duration": 178.91206666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a baby being fed from a bottle in black and white?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a baby being fed from a bottle in black and white?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 29.0], "base_se_norm": [-0.7205331325531006, -0.6758183836936951], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.7205331325531006, -0.6758183836936951], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004049.pt"}
{"idx": 4051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksdOh7MbRgk.mp4", "duration": 178.91206666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where two young girls are playing and laughing together in a swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where two young girls are playing and laughing together in a swimming pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 77.0], "base_se_norm": [-0.18395662307739258, -0.13924193382263184], "gt_se_raw": [73.0, 77.0], "gt_se_norm": [-0.18395662307739258, -0.13924193382263184], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004051.pt"}
{"idx": 4053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksdOh7MbRgk.mp4", "duration": 178.91206666666668, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range where a boy is sitting on a park bench with a blue Braille typewriter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range where a boy is sitting on a park bench with a blue Braille typewriter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 142.0], "base_se_norm": [0.5426572561264038, 0.5873719453811646], "gt_se_raw": [138.0, 141.0], "gt_se_norm": [0.5426572561264038, 0.5761933326721191], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004053.pt"}
{"idx": 4055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFjp1MW6Lzc.mp4", "duration": 119.96, "prompt_text": "Please find the visual event described by the sentence 'When does the boy secure the newspapers onto the back of his bike?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy secure the newspapers onto the back of his bike?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 39.0], "base_se_norm": [-0.44981658458709717, -0.3497832417488098], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.43314439058303833, -0.3664554953575134], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004055.pt"}
{"idx": 4057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFjp1MW6Lzc.mp4", "duration": 119.96, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy show the newspaper to the blind man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy show the newspaper to the blind man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 96.0], "base_se_norm": [0.5338445901870728, 0.6005334854125977], "gt_se_raw": [96.0, 103.0], "gt_se_norm": [0.6005334854125977, 0.7172391414642334], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004057.pt"}
{"idx": 4059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CVtTOpgSyY.mp4", "duration": 30.833333333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the boy pull the anchor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy pull the anchor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.6108108162879944], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9351351261138916, -0.6108108162879944], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004059.pt"}
{"idx": 4061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CVtTOpgSyY.mp4", "duration": 30.833333333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does a shooting star streak across the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a shooting star streak across the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [-0.02702707052230835, 0.10270273685455322], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.02702707052230835, 0.10270273685455322], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004061.pt"}
{"idx": 4063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7CVtTOpgSyY.mp4", "duration": 30.833333333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy squats down to examine a glowing star on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy squats down to examine a glowing star on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 29.0], "base_se_norm": [0.42702698707580566, 0.8810809850692749], "gt_se_raw": [22.0, 29.0], "gt_se_norm": [0.42702698707580566, 0.8810809850692749], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004063.pt"}
{"idx": 4065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRi80V8ire8.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman is hugging her black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman is hugging her black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.7752248048782349, -0.7252747416496277], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.8001998066902161, -0.7502497434616089], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004065.pt"}
{"idx": 4067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GRi80V8ire8.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is making pottery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is making pottery.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 51.0], "base_se_norm": [0.17382609844207764, 0.273726224899292], "gt_se_raw": [47.0, 51.0], "gt_se_norm": [0.17382609844207764, 0.273726224899292], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004067.pt"}
{"idx": 4069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xWggTb45brM.mp4", "duration": 311.5, "prompt_text": "Please find the visual event described by the sentence 'At what time can the dome-shaped stadium be seen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time can the dome-shaped stadium be seen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 9.0], "base_se_norm": [-0.9871588945388794, -0.942215085029602], "gt_se_raw": [2.0, 9.0], "gt_se_norm": [-0.9871588945388794, -0.942215085029602], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004069.pt"}
{"idx": 4071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xWggTb45brM.mp4", "duration": 311.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man dances in front of the grand piano.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man dances in front of the grand piano.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 112.0], "base_se_norm": [-0.39004814624786377, -0.2808988690376282], "gt_se_raw": [98.0, 111.0], "gt_se_norm": [-0.37078648805618286, -0.2873194217681885], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004071.pt"}
{"idx": 4073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xWggTb45brM.mp4", "duration": 311.5, "prompt_text": "Please find the visual event described by the sentence 'Show me the aerial shot of the house with fireworks at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 276.0 - 301.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the aerial shot of the house with fireworks at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 276.0 - 301.0 seconds.<|im_end|>\n", "base_se_raw": [276.0, 301.0], "base_se_norm": [0.7720706462860107, 0.932584285736084], "gt_se_raw": [277.0, 291.0], "gt_se_norm": [0.7784911394119263, 0.868378758430481], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004073.pt"}
{"idx": 4075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/81kd2SbB0qM.mp4", "duration": 31.5315, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy in the walker move past the first white door on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy in the walker move past the first white door on the left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9365713596343994, -0.3657136559486389], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5559995174407959, -0.11199909448623657], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004075.pt"}
{"idx": 4077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/81kd2SbB0qM.mp4", "duration": 31.5315, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman briefly pats the boy on his back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman briefly pats the boy on his back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.048570454120635986, 0.141715407371521], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.141715407371521, 0.26857268810272217], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004077.pt"}
{"idx": 4079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4dgJoFwlhE.mp4", "duration": 113.48003333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the people wearing hard hats and safety vests?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the people wearing hard hats and safety vests?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 77.0], "base_se_norm": [0.321818470954895, 0.3570668697357178], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.321818470954895, 0.3746911287307739], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004079.pt"}
{"idx": 4081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4dgJoFwlhE.mp4", "duration": 113.48003333333334, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a man walking down a corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a man walking down a corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.823757529258728, -0.7885090708732605], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.823757529258728, -0.7885090708732605], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004081.pt"}
{"idx": 4083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n4dgJoFwlhE.mp4", "duration": 113.48003333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the video show people having fun on a beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show people having fun on a beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.9647514820098877, -0.8766303062438965], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.9647514820098877, -0.8766303062438965], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004083.pt"}
{"idx": 4085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oM_M_d9OiHc.mp4", "duration": 75.3, "prompt_text": "Please find the visual event described by the sentence 'At what moment does a lifeguard explain the precautions taken at the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does a lifeguard explain the precautions taken at the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 16.0], "base_se_norm": [-0.7343957424163818, -0.5750331878662109], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.7609561681747437, -0.5750331878662109], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004085.pt"}
{"idx": 4087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oM_M_d9OiHc.mp4", "duration": 75.3, "prompt_text": "Please find the visual event described by the sentence 'When does a person jump into the water despite the warnings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a person jump into the water despite the warnings?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 47.0], "base_se_norm": [0.16865861415863037, 0.24833989143371582], "gt_se_raw": [44.0, 47.0], "gt_se_norm": [0.16865861415863037, 0.24833989143371582], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004087.pt"}
{"idx": 4089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j4vo5y0eHnA.mp4", "duration": 81.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the adults and children lifting the parachute together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the adults and children lifting the parachute together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8770492076873779, -0.7540983557701111], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.8770492076873779, -0.7786885499954224], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004089.pt"}
{"idx": 4091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j4vo5y0eHnA.mp4", "duration": 81.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'A woman brings a child to the front of the parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman brings a child to the front of the parachute.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 63.0], "base_se_norm": [0.008196711540222168, 0.549180269241333], "gt_se_raw": [44.0, 50.0], "gt_se_norm": [0.08196723461151123, 0.2295081615447998], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004091.pt"}
{"idx": 4093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j4vo5y0eHnA.mp4", "duration": 81.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'When do all the children run out from under the parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do all the children run out from under the parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 80.0], "base_se_norm": [0.7950818538665771, 0.9672130346298218], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [0.7950818538665771, 0.9180327653884888], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004093.pt"}
{"idx": 4095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5N2BC6PX3Us.mp4", "duration": 141.8, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the title card 'YOU CAN'T ASK THAT' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the title card 'YOU CAN'T ASK THAT' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 31.0], "base_se_norm": [-0.6473907232284546, -0.5627644658088684], "gt_se_raw": [24.0, 31.0], "gt_se_norm": [-0.6614950895309448, -0.5627644658088684], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004095.pt"}
{"idx": 4097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5N2BC6PX3Us.mp4", "duration": 141.8, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man is energetically signing on an outdoor stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man is energetically signing on an outdoor stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 100.0], "base_se_norm": [0.2693934440612793, 0.4104372262954712], "gt_se_raw": [90.0, 100.0], "gt_se_norm": [0.2693934440612793, 0.4104372262954712], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004097.pt"}
{"idx": 4099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iWqNo5DnjtA.mp4", "duration": 107.34056666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man begin to slice the ginger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man begin to slice the ginger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 58.0], "base_se_norm": [-0.12428265810012817, 0.0806725025177002], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.08701807260513306, -0.03112119436264038], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004099.pt"}
{"idx": 4101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iWqNo5DnjtA.mp4", "duration": 107.34056666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a plate of pulled pork sliders with coleslaw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a plate of pulled pork sliders with coleslaw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9627354145050049], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.9813677072525024], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004101.pt"}
{"idx": 4103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iWqNo5DnjtA.mp4", "duration": 107.34056666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man introduce himself and the topic of mincing ginger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man introduce himself and the topic of mincing ginger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 38.0], "base_se_norm": [-0.8695740103721619, -0.29197317361831665], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.8695740103721619, -0.7205157279968262], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004103.pt"}
{"idx": 4105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1P6zkiJ6EQA.mp4", "duration": 60.76067734401068, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a man signs a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a man signs a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [0.6128852367401123, 0.7116333246231079], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [0.645801305770874, 0.7445493936538696], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004105.pt"}
{"idx": 4107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1P6zkiJ6EQA.mp4", "duration": 60.76067734401068, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when a young man picks up a laptop and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when a young man picks up a laptop and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 58.0], "base_se_norm": [0.8103814125061035, 0.9091295003890991], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [0.8103814125061035, 0.9091295003890991], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004107.pt"}
{"idx": 4109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4sx42_C10zw.mp4", "duration": 105.0, "prompt_text": "Please find the visual event described by the sentence 'When is the phrase 'Writing an Effective Thesis' being written on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the phrase 'Writing an Effective Thesis' being written on the whiteboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9428571462631226], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9428571462631226], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004109.pt"}
{"idx": 4111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4sx42_C10zw.mp4", "duration": 105.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show the example about 'Stress in the fast-food workplace'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show the example about 'Stress in the fast-food workplace'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 68.0], "base_se_norm": [-0.02857142686843872, 0.2952381372451782], "gt_se_raw": [51.0, 58.0], "gt_se_norm": [-0.02857142686843872, 0.10476195812225342], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004111.pt"}
{"idx": 4113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4sx42_C10zw.mp4", "duration": 105.0, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a tombstone with 'RIP' is being drawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a tombstone with 'RIP' is being drawn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 93.0], "base_se_norm": [0.6952381134033203, 0.7714285850524902], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [0.6952381134033203, 0.7523809671401978], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004113.pt"}
{"idx": 4115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IPPs6cIFZqM.mp4", "duration": 491.62446666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman drinks the whole glass of liquor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman drinks the whole glass of liquor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 74.0], "base_se_norm": [-0.7355705499649048, -0.6989572048187256], "gt_se_raw": [66.0, 74.0], "gt_se_norm": [-0.7315024137496948, -0.6989572048187256], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004115.pt"}
{"idx": 4117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IPPs6cIFZqM.mp4", "duration": 491.62446666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is applying blush with a large red brush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 257.0 - 268.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is applying blush with a large red brush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 257.0 - 268.0 seconds.<|im_end|>\n", "base_se_raw": [257.0, 268.0], "base_se_norm": [0.04551339149475098, 0.0902630090713501], "gt_se_raw": [289.0, 302.0], "gt_se_norm": [0.17569410800933838, 0.22857999801635742], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004117.pt"}
{"idx": 4119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q6rAllJAdWk.mp4", "duration": 176.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man press the red button to place an order?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man press the red button to place an order?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.9773585200309753, -0.9320755004882812], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9660377502441406, -0.9433962106704712], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004119.pt"}
{"idx": 4121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q6rAllJAdWk.mp4", "duration": 176.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man give a tip to the employee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man give a tip to the employee?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.7849056720733643, -0.705660343170166], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.7735849022865295, -0.7396226525306702], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004121.pt"}
{"idx": 4123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q6rAllJAdWk.mp4", "duration": 176.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man starts drinking the strawberry shake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man starts drinking the strawberry shake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 122.0], "base_se_norm": [0.14339613914489746, 0.38113200664520264], "gt_se_raw": [102.0, 105.0], "gt_se_norm": [0.15471696853637695, 0.18867921829223633], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004123.pt"}
{"idx": 4125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6yYd6Pq7Ic.mp4", "duration": 64.25, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'World's Best Cup of Coffee' sign displayed on the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'World's Best Cup of Coffee' sign displayed on the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.4708171486854553, -0.3463035225868225], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [-0.4708171486854553, -0.3463035225868225], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004125.pt"}
{"idx": 4127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6yYd6Pq7Ic.mp4", "duration": 64.25, "prompt_text": "Please find the visual event described by the sentence 'When does the man dressed as an elf walk with a man in a red jumpsuit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man dressed as an elf walk with a man in a red jumpsuit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [0.4319065809249878, 0.618677020072937], "gt_se_raw": [46.0, 51.0], "gt_se_norm": [0.4319065809249878, 0.5875486135482788], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004127.pt"}
{"idx": 4129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r7MWpMelEyg.mp4", "duration": 114.58113333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the child focused on getting the contents out of the clear shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the child focused on getting the contents out of the clear shaker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 30.0], "base_se_norm": [-0.9476353526115417, -0.47635358572006226], "gt_se_raw": [4.0, 30.0], "gt_se_norm": [-0.9301804900169373, -0.47635358572006226], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004129.pt"}
{"idx": 4131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r7MWpMelEyg.mp4", "duration": 114.58113333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl pick up and examine the pink object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl pick up and examine the pink object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 67.0], "base_se_norm": [0.029837965965270996, 0.16947698593139648], "gt_se_raw": [63.0, 68.0], "gt_se_norm": [0.09965753555297852, 0.18693184852600098], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004131.pt"}
{"idx": 4133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r7MWpMelEyg.mp4", "duration": 114.58113333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl playing with the jingle bells on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl playing with the jingle bells on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 106.0], "base_se_norm": [0.536029577255249, 0.850217342376709], "gt_se_raw": [89.0, 104.0], "gt_se_norm": [0.5534844398498535, 0.8153076171875], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004133.pt"}
{"idx": 4135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/egkrxkiUnoo.mp4", "duration": 150.01653333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the pink shirt sit down next to the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the pink shirt sit down next to the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 17.0], "base_se_norm": [-0.8266857862472534, -0.7733583450317383], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.8266857862472534, -0.7733583450317383], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004135.pt"}
{"idx": 4137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/egkrxkiUnoo.mp4", "duration": 150.01653333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man points at the woman's shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man points at the woman's shoes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.24008381366729736, -0.17342448234558105], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [-0.24008381366729736, -0.20008820295333862], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004137.pt"}
{"idx": 4139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qi2m4V21bw4.mp4", "duration": 179.0, "prompt_text": "Please find the visual event described by the sentence 'When is Mr. Bean preparing pasta in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Mr. Bean preparing pasta in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 36.0], "base_se_norm": [-0.9329608678817749, -0.597765326499939], "gt_se_raw": [3.0, 36.0], "gt_se_norm": [-0.9664804339408875, -0.597765326499939], "start_token_span": [71, 74], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004139.pt"}
{"idx": 4141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qi2m4V21bw4.mp4", "duration": 179.0, "prompt_text": "Please find the visual event described by the sentence 'When does the woman at the table eat the flower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman at the table eat the flower?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 88.0], "base_se_norm": [-0.061452507972717285, -0.016759753227233887], "gt_se_raw": [79.0, 88.0], "gt_se_norm": [-0.11731845140457153, -0.016759753227233887], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004141.pt"}
{"idx": 4143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qi2m4V21bw4.mp4", "duration": 179.0, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a bird steals bread through the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a bird steals bread through the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 124.0], "base_se_norm": [0.27374303340911865, 0.3854748010635376], "gt_se_raw": [115.0, 118.0], "gt_se_norm": [0.2849161624908447, 0.31843578815460205], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004143.pt"}
{"idx": 4145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LKFuXETZUsI.mp4", "duration": 154.40425, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Moana and Maui are on the boat together at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Moana and Maui are on the boat together at sunset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 78.0], "base_se_norm": [-0.18396031856536865, 0.01033484935760498], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.18396031856536865, -0.15805429220199585], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004145.pt"}
{"idx": 4147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LKFuXETZUsI.mp4", "duration": 154.40425, "prompt_text": "Please find the visual event described by the sentence 'Where do we see the little coconut creatures for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where do we see the little coconut creatures for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.9481879472732544, -0.9222819209098816], "gt_se_raw": [83.0, 85.0], "gt_se_norm": [0.07509994506835938, 0.1010059118270874], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004147.pt"}
{"idx": 4149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PjCZVeR0do8.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator mention that the employment rate for people without disabilities is significantly higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator mention that the employment rate for people without disabilities is significantly higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 6.0], "base_se_norm": [-0.9031662940979004, -0.4189978241920471], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9031662940979004, -0.4189978241920471], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004149.pt"}
{"idx": 4151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WNHJnUIGQ_U.mp4", "duration": 95.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman using a white comb on the girl's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman using a white comb on the girl's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 17.0], "base_se_norm": [-1.0, -0.6438547372817993], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.2248603105545044, -0.14106148481369019], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004151.pt"}
{"idx": 4153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WNHJnUIGQ_U.mp4", "duration": 95.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the girl smells the contents of the yellow bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the girl smells the contents of the yellow bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 67.0], "base_se_norm": [0.1312849521636963, 0.403631329536438], "gt_se_raw": [85.0, 88.0], "gt_se_norm": [0.7807261943817139, 0.8435753583908081], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004153.pt"}
{"idx": 4155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WNHJnUIGQ_U.mp4", "duration": 95.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman says 'Let's brush our teeth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman says 'Let's brush our teeth'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 22.0], "base_se_norm": [-0.6438547372817993, -0.5391061305999756], "gt_se_raw": [94.0, 95.0], "gt_se_norm": [0.9692736864089966, 0.9902234077453613], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004155.pt"}
{"idx": 4157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_lmKuKrsKRA.mp4", "duration": 82.457375, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man takes his first bite of a pancake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man takes his first bite of a pancake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.36937111616134644, -0.22384142875671387], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.36937111616134644, -0.32086122035980225], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004157.pt"}
{"idx": 4159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_lmKuKrsKRA.mp4", "duration": 82.457375, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the woman hands the plate of pancakes to the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the woman hands the plate of pancakes to the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.0055468082427978516, 0.06721806526184082], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.0055468082427978516, 0.04296314716339111], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004159.pt"}
{"idx": 4161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ppyYdn2nPoU.mp4", "duration": 124.72472472472472, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the guest speaker is giving a speech during the assembly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the guest speaker is giving a speech during the assembly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.7755056023597717, -0.7273997068405151], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [-0.7755056023597717, -0.7273997068405151], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004161.pt"}
{"idx": 4163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ppyYdn2nPoU.mp4", "duration": 124.72472472472472, "prompt_text": "Please find the visual event described by the sentence 'At what point are people participating in the scavenger hunt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are people participating in the scavenger hunt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 66.0], "base_se_norm": [-0.31048154830932617, 0.058330655097961426], "gt_se_raw": [44.0, 66.0], "gt_se_norm": [-0.2944462299346924, 0.058330655097961426], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004163.pt"}
{"idx": 4165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ppyYdn2nPoU.mp4", "duration": 124.72472472472472, "prompt_text": "Please find the visual event described by the sentence 'When can we see people participating in the walk-a-thon by the lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see people participating in the walk-a-thon by the lake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 118.0], "base_se_norm": [0.5233547687530518, 0.8921669721603394], "gt_se_raw": [89.0, 118.0], "gt_se_norm": [0.427142858505249, 0.8921669721603394], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004165.pt"}
{"idx": 4167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/16K6BDTn9iI.mp4", "duration": 28.72, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men dueling with swords on a large circular mosaic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men dueling with swords on a large circular mosaic?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.025069594383239746, 0.11420619487762451], "gt_se_raw": [15.0, 16.0], "gt_se_norm": [0.04456830024719238, 0.11420619487762451], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004167.pt"}
{"idx": 4169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/16K6BDTn9iI.mp4", "duration": 28.72, "prompt_text": "Please find the visual event described by the sentence 'At what point does a running skeleton appear in a snowy environment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a running skeleton appear in a snowy environment?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.6016713380813599, 0.7409471273422241], "gt_se_raw": [23.0, 24.0], "gt_se_norm": [0.6016713380813599, 0.671309232711792], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004169.pt"}
{"idx": 4171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XbKhZYOvx8.mp4", "duration": 83.08, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a suit and tie talking to a group of people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a suit and tie talking to a group of people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.5426095724105835, -0.4944632053375244], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.5666826963424683, -0.5185363292694092], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004171.pt"}
{"idx": 4173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XbKhZYOvx8.mp4", "duration": 83.08, "prompt_text": "Please find the visual event described by the sentence 'When are the protestors posing for a photo in front of a Nike store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the protestors posing for a photo in front of a Nike store?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [0.13143956661224365, 0.2518054246902466], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [0.15551269054412842, 0.22773230075836182], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004173.pt"}
{"idx": 4175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XbKhZYOvx8.mp4", "duration": 83.08, "prompt_text": "Please find the visual event described by the sentence 'Identify the timeframe where a person is holding a red and white umbrella.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the timeframe where a person is holding a red and white umbrella.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 31.0], "base_se_norm": [-0.3018777370452881, -0.253731369972229], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.3018777370452881, -0.253731369972229], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004175.pt"}
{"idx": 4177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xqZQ9KM_LjY.mp4", "duration": 198.7986, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man in the black shirt is enjoying the imaginary ribs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man in the black shirt is enjoying the imaginary ribs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 64.0], "base_se_norm": [-0.44667619466781616, -0.35613226890563965], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.45673662424087524, -0.426555335521698], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004177.pt"}
{"idx": 4179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xqZQ9KM_LjY.mp4", "duration": 198.7986, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a bikini caressing a man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a bikini caressing a man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 114.0], "base_se_norm": [0.0965871810913086, 0.146889328956604], "gt_se_raw": [109.0, 112.0], "gt_se_norm": [0.0965871810913086, 0.12676846981048584], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004179.pt"}
{"idx": 4181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jHe5vPlKgJA.mp4", "duration": 21.755066666666668, "prompt_text": "Please find the visual event described by the sentence 'When are the two boys sitting on the couch and looking at the device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two boys sitting on the couch and looking at the device?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, 0.3789891004562378], "gt_se_raw": [1.0, 5.0], "gt_se_norm": [-0.9080674052238464, -0.5403369665145874], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004181.pt"}
{"idx": 4183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jHe5vPlKgJA.mp4", "duration": 21.755066666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy in the gray shirt point at the screen for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy in the gray shirt point at the screen for the second time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.17260652780532837, 0.011258721351623535], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.08067387342453003, 0.011258721351623535], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004183.pt"}
{"idx": 4185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jHe5vPlKgJA.mp4", "duration": 21.755066666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where one boy is carrying the other on his back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where one boy is carrying the other on his back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 20.0], "base_se_norm": [0.47092175483703613, 0.8386522531509399], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [0.6547869443893433, 0.8386522531509399], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004185.pt"}
{"idx": 4187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2bpkd6hwH6U.mp4", "duration": 143.768625, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman falls into the tank of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman falls into the tank of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.596574068069458, -0.5409290790557861], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.596574068069458, -0.5687515735626221], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004187.pt"}
{"idx": 4189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2bpkd6hwH6U.mp4", "duration": 143.768625, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the tank full of piranhas is dropped into the water below.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the tank full of piranhas is dropped into the water below.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 83.0], "base_se_norm": [0.05725431442260742, 0.15463292598724365], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [0.07116556167602539, 0.09898805618286133], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004189.pt"}
{"idx": 4191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uidTVaoTjNY.mp4", "duration": 130.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the two brothers shake hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two brothers shake hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.9079519510269165, -0.8772692680358887], "gt_se_raw": [7.0, 8.0], "gt_se_norm": [-0.8926106095314026, -0.8772692680358887], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004191.pt"}
{"idx": 4193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uidTVaoTjNY.mp4", "duration": 130.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the three orange circles listing benefits displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three orange circles listing benefits displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 77.0], "base_se_norm": [0.08923542499542236, 0.1812835931777954], "gt_se_raw": [71.0, 77.0], "gt_se_norm": [0.08923542499542236, 0.1812835931777954], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004193.pt"}
{"idx": 4195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uidTVaoTjNY.mp4", "duration": 130.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the list of benefits for employing people with disabilities appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the list of benefits for employing people with disabilities appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 99.0], "base_se_norm": [0.42674505710601807, 0.5187931060791016], "gt_se_raw": [92.0, 99.0], "gt_se_norm": [0.4114036560058594, 0.5187931060791016], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004195.pt"}
{"idx": 4197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mEBtnA4DIM4.mp4", "duration": 123.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man in the suit checks his briefcase for a phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man in the suit checks his briefcase for a phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.7254374027252197, -0.6608344316482544], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.5962314605712891, -0.5477793216705322], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004197.pt"}
{"idx": 4199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mEBtnA4DIM4.mp4", "duration": 123.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man arrives and says he will fix the escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man arrives and says he will fix the escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 20.0], "base_se_norm": [-0.9676985144615173, -0.6769852042198181], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [0.14670252799987793, 0.21130549907684326], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004199.pt"}
{"idx": 4201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V99PmrBPUqk.mp4", "duration": 161.28, "prompt_text": "Please find the visual event described by the sentence 'When is the presenter introducing the topic at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the presenter introducing the topic at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 12.0], "base_se_norm": [-0.9627976417541504, -0.851190447807312], "gt_se_raw": [3.0, 12.0], "gt_se_norm": [-0.9627976417541504, -0.851190447807312], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004201.pt"}
{"idx": 4203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V99PmrBPUqk.mp4", "duration": 161.28, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the table showing statistics on wheelchair users?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the table showing statistics on wheelchair users?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 58.0], "base_se_norm": [-0.6403769850730896, -0.2807539701461792], "gt_se_raw": [29.0, 58.0], "gt_se_norm": [-0.6403769850730896, -0.2807539701461792], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004203.pt"}
{"idx": 4205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V99PmrBPUqk.mp4", "duration": 161.28, "prompt_text": "Please find the visual event described by the sentence 'When is the driver shown using the joystick to operate the vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the driver shown using the joystick to operate the vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 122.0], "base_se_norm": [0.4384920597076416, 0.5128968954086304], "gt_se_raw": [116.0, 122.0], "gt_se_norm": [0.4384920597076416, 0.5128968954086304], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004205.pt"}
{"idx": 4207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsQMSvhMcMc.mp4", "duration": 124.39093333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video show the webpage for booking timed-entry passes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video show the webpage for booking timed-entry passes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 35.0], "base_se_norm": [-0.5658847093582153, -0.4372580051422119], "gt_se_raw": [27.0, 36.0], "gt_se_norm": [-0.5658847093582153, -0.4211796522140503], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004207.pt"}
{"idx": 4209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KsQMSvhMcMc.mp4", "duration": 124.39093333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the video display the museum's floor plan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video display the museum's floor plan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 88.0], "base_se_norm": [0.28626739978790283, 0.41489410400390625], "gt_se_raw": [79.0, 89.0], "gt_se_norm": [0.2701890468597412, 0.43097245693206787], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004209.pt"}
{"idx": 4211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cda7mSowTEI.mp4", "duration": 75.50884217550885, "prompt_text": "Please find the visual event described by the sentence 'When is the university president speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the university president speaking to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 17.0], "base_se_norm": [-0.9470260739326477, -0.5497215986251831], "gt_se_raw": [2.0, 17.0], "gt_se_norm": [-0.9470260739326477, -0.5497215986251831], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004211.pt"}
{"idx": 4213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cda7mSowTEI.mp4", "duration": 75.50884217550885, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the instructor is demonstrating bookmaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the instructor is demonstrating bookmaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 45.0], "base_se_norm": [0.0594785213470459, 0.19191336631774902], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [0.08596551418304443, 0.21840035915374756], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004213.pt"}
{"idx": 4215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cda7mSowTEI.mp4", "duration": 75.50884217550885, "prompt_text": "Please find the visual event described by the sentence 'Where in the video is there a tutorial on how to add comments to an image?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video is there a tutorial on how to add comments to an image?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 70.0], "base_se_norm": [0.721652626991272, 0.8540874719619751], "gt_se_raw": [65.0, 70.0], "gt_se_norm": [0.721652626991272, 0.8540874719619751], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004215.pt"}
{"idx": 4217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jhfLlamufKE.mp4", "duration": 56.72333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the food slides out of the can and onto the plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the food slides out of the can and onto the plate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.012751936912536621, 0.12828350067138672], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [-0.012751936912536621, 0.057765841484069824], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004217.pt"}
{"idx": 4219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jhfLlamufKE.mp4", "duration": 56.72333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the can lid with a smiley face pop off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the can lid with a smiley face pop off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 30.0], "base_se_norm": [-0.012751936912536621, 0.057765841484069824], "gt_se_raw": [28.0, 29.0], "gt_se_norm": [-0.012751936912536621, 0.0225069522857666], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004219.pt"}
{"idx": 4221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g26mbST0YhU.mp4", "duration": 91.4914, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman pushing the wheelchair forward?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman pushing the wheelchair forward?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.781400203704834], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9125601053237915], "start_token_span": [71, 74], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004221.pt"}
{"idx": 4223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g26mbST0YhU.mp4", "duration": 91.4914, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman bend over to adjust something on the boy's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman bend over to adjust something on the boy's lap?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 62.0], "base_se_norm": [0.11485886573791504, 0.35531866550445557], "gt_se_raw": [51.0, 61.0], "gt_se_norm": [0.11485886573791504, 0.33345866203308105], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004223.pt"}
{"idx": 4225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g26mbST0YhU.mp4", "duration": 91.4914, "prompt_text": "Please find the visual event described by the sentence 'When is the boy moving the walking aid from side to side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy moving the walking aid from side to side?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 67.0], "base_se_norm": [-0.16932082176208496, 0.4646185636520386], "gt_se_raw": [39.0, 46.0], "gt_se_norm": [-0.14746087789535522, 0.005558967590332031], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004225.pt"}
{"idx": 4227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uNefa-pBVD0.mp4", "duration": 75.04163333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see two people boxing with each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two people boxing with each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.2803994417190552, -0.14714008569717407], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.2803994417190552, -0.14714008569717407], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004227.pt"}
{"idx": 4229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uNefa-pBVD0.mp4", "duration": 75.04163333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip of the classroom installation with golden 'BLACK' and 'WHITE' signs on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip of the classroom installation with golden 'BLACK' and 'WHITE' signs on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [0.25263798236846924, 0.4925048351287842], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [0.38589727878570557, 0.5191566944122314], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004229.pt"}
{"idx": 4231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7cSIiKxEt4.mp4", "duration": 70.87078745412079, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman introduces herself as the president of the university.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman introduces herself as the president of the university.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 22.0], "base_se_norm": [-0.9435592293739319, -0.3791518211364746], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.9435592293739319, -0.8588981628417969], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004231.pt"}
{"idx": 4233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7cSIiKxEt4.mp4", "duration": 70.87078745412079, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the audio mixing board with its lights on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the audio mixing board with its lights on.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.012286961078643799, 0.07237410545349121], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.012286961078643799, 0.07237410545349121], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004233.pt"}
{"idx": 4235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b7cSIiKxEt4.mp4", "duration": 70.87078745412079, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man with headphones speak into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man with headphones speak into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 50.0], "base_se_norm": [0.354577898979187, 0.41101861000061035], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [0.7778834104537964, 0.8625445365905762], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004235.pt"}
{"idx": 4237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zYVR5PDom8.mp4", "duration": 88.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where two girls are drawing together at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where two girls are drawing together at a table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 38.0], "base_se_norm": [-0.4572182297706604, -0.14059555530548096], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.4346023201942444, -0.208443284034729], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004237.pt"}
{"idx": 4239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zYVR5PDom8.mp4", "duration": 88.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a girl asks a boy if she can use the paddle ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a girl asks a boy if she can use the paddle ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 66.0], "base_se_norm": [0.3117225170135498, 0.49264979362487793], "gt_se_raw": [60.0, 66.0], "gt_se_norm": [0.35695433616638184, 0.49264979362487793], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004239.pt"}
{"idx": 4241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ENaJcHwQEVs.mp4", "duration": 56.62323333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the boy moving his feet up and down the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy moving his feet up and down the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 33.0], "base_se_norm": [-0.5408245325088501, 0.16559922695159912], "gt_se_raw": [13.0, 34.0], "gt_se_norm": [-0.5408245325088501, 0.2009204626083374], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004241.pt"}
{"idx": 4243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ENaJcHwQEVs.mp4", "duration": 56.62323333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is the boy seen playing with a blue stringy object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the boy seen playing with a blue stringy object?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 56.0], "base_se_norm": [0.412847638130188, 0.9779866933822632], "gt_se_raw": [40.0, 56.0], "gt_se_norm": [0.412847638130188, 0.9779866933822632], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004243.pt"}
{"idx": 4245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ENaJcHwQEVs.mp4", "duration": 56.62323333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where the boy is holding the blue item above his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where the boy is holding the blue item above his head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 54.0], "base_se_norm": [0.412847638130188, 0.9073442220687866], "gt_se_raw": [50.0, 54.0], "gt_se_norm": [0.7660595178604126, 0.9073442220687866], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004245.pt"}
{"idx": 4247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UGvlQOWyB2Y.mp4", "duration": 31.533333333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a person in a mascot costume is spinning and waving a flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a person in a mascot costume is spinning and waving a flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.746300220489502, -0.3657504916191101], "gt_se_raw": [5.0, 9.0], "gt_se_norm": [-0.6828752756118774, -0.4291754961013794], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004247.pt"}
{"idx": 4249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UGvlQOWyB2Y.mp4", "duration": 31.533333333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the mascot is waving a flag near the boards on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the mascot is waving a flag near the boards on the right side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.04862576723098755, 0.3319239616394043], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.04862576723098755, 0.2684990167617798], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004249.pt"}
{"idx": 4251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDvBqq3v8bM.mp4", "duration": 360.89386666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man with a prosthetic leg and a cane standing alone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man with a prosthetic leg and a cane standing alone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 134.0], "base_se_norm": [-0.296191930770874, -0.25739938020706177], "gt_se_raw": [127.0, 134.0], "gt_se_norm": [-0.296191930770874, -0.25739938020706177], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004251.pt"}
{"idx": 4253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TDvBqq3v8bM.mp4", "duration": 360.89386666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where people are learning how to plant saplings in a workshop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 277.0 - 284.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where people are learning how to plant saplings in a workshop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 277.0 - 284.0 seconds.<|im_end|>\n", "base_se_raw": [277.0, 284.0], "base_se_norm": [0.5350773334503174, 0.5738699436187744], "gt_se_raw": [278.0, 287.0], "gt_se_norm": [0.5406191349029541, 0.5904953479766846], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004253.pt"}
{"idx": 4255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7VS1wPeWqAA.mp4", "duration": 83.5835, "prompt_text": "Please find the visual event described by the sentence 'When is the man sitting alone at the table before two women enter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man sitting alone at the table before two women enter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.8325028419494629], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.8325028419494629], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004255.pt"}
{"idx": 4257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7VS1wPeWqAA.mp4", "duration": 83.5835, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the black dress say goodbye and walk away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the black dress say goodbye and walk away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.018945157527923584, 0.05283927917480469], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.0428733229637146, 0.028911113739013672], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004257.pt"}
{"idx": 4259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7VS1wPeWqAA.mp4", "duration": 83.5835, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man put his head in his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man put his head in his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 64.0], "base_se_norm": [0.4356900453567505, 0.5314027070999146], "gt_se_raw": [61.0, 64.0], "gt_se_norm": [0.4596182107925415, 0.5314027070999146], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004259.pt"}
{"idx": 4261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM7Mh9T_XE.mp4", "duration": 142.35053333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point do the man and woman start watching cartoons on TV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the man and woman start watching cartoons on TV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 57.0], "base_se_norm": [-0.35370808839797974, -0.19916003942489624], "gt_se_raw": [48.0, 54.0], "gt_se_norm": [-0.32560843229293823, -0.2413095235824585], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004261.pt"}
{"idx": 4263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xhM7Mh9T_XE.mp4", "duration": 142.35053333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the plaid shirt open the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the plaid shirt open the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 23.0], "base_se_norm": [-0.8033024668693542, -0.6768540143966675], "gt_se_raw": [13.0, 24.0], "gt_se_norm": [-0.817352294921875, -0.6628042459487915], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004263.pt"}
{"idx": 4265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cHUNbTfzOr4.mp4", "duration": 72.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman introducing herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman introducing herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.7240110635757446], "gt_se_raw": [2.0, 6.0], "gt_se_norm": [-0.9448022246360779, -0.8344066143035889], "start_token_span": [67, 70], "end_token_span": [66, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004265.pt"}
{"idx": 4267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cHUNbTfzOr4.mp4", "duration": 72.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is explaining her job responsibilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is explaining her job responsibilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 50.0], "base_se_norm": [-0.9724010825157166, 0.3799448013305664], "gt_se_raw": [12.0, 26.0], "gt_se_norm": [-0.6688132286071777, -0.28242868185043335], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004267.pt"}
{"idx": 4269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cHUNbTfzOr4.mp4", "duration": 72.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the part of the video that shows the RAMP logo and contact phone numbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part of the video that shows the RAMP logo and contact phone numbers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 72.0], "base_se_norm": [0.766329288482666, 0.987120509147644], "gt_se_raw": [64.0, 72.0], "gt_se_norm": [0.766329288482666, 0.987120509147644], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004269.pt"}
{"idx": 4271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kgDjwR0xoQg.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the speaker explains that olfactory information gets to the brain first.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the speaker explains that olfactory information gets to the brain first.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [0.05077850818634033, 0.1878366470336914], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.4060816764831543, -0.3603956699371338], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004271.pt"}
{"idx": 4273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kgDjwR0xoQg.mp4", "duration": 87.55413333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker gesture with both hands while explaining how other senses 'catch up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker gesture with both hands while explaining how other senses 'catch up'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 41.0], "base_se_norm": [-0.3147096633911133, -0.06343650817871094], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [0.1193075180053711, 0.1649935245513916], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004273.pt"}
{"idx": 4275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wC1ZQNR_1N4.mp4", "duration": 192.5, "prompt_text": "Please find the visual event described by the sentence 'When does the employee hand the food and drinks to the customer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the employee hand the food and drinks to the customer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9896103739738464, -0.864935040473938], "gt_se_raw": [1.0, 13.0], "gt_se_norm": [-0.9896103739738464, -0.864935040473938], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004275.pt"}
{"idx": 4277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wC1ZQNR_1N4.mp4", "duration": 192.5, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man remove the lid from his cup to show the drink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 61.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man remove the lid from his cup to show the drink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 61.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [61.0, 65.0], "base_se_norm": [-0.366233766078949, -0.3246753215789795], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.3454545736312866, -0.3246753215789795], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004277.pt"}
{"idx": 4279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wC1ZQNR_1N4.mp4", "duration": 192.5, "prompt_text": "Please find the visual event described by the sentence 'When does the man react to getting a brain freeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man react to getting a brain freeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 147.0], "base_se_norm": [0.41298699378967285, 0.5272727012634277], "gt_se_raw": [137.0, 145.0], "gt_se_norm": [0.4233766794204712, 0.5064934492111206], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004279.pt"}
{"idx": 4281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OxrBik16Hzg.mp4", "duration": 95.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'When do the police officers in beige uniforms start to intervene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the police officers in beige uniforms start to intervene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 30.0], "base_se_norm": [-0.5594406127929688, -0.37062937021255493], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.5594406127929688, -0.4965035319328308], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004281.pt"}
{"idx": 4283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OxrBik16Hzg.mp4", "duration": 95.33333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where monks are kneeling and praying in front of the burning man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where monks are kneeling and praying in front of the burning man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 72.0], "base_se_norm": [-0.37062937021255493, 0.5104894638061523], "gt_se_raw": [71.0, 74.0], "gt_se_norm": [0.4895104169845581, 0.5524475574493408], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004283.pt"}
{"idx": 4285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5VmfgTfuYGE.mp4", "duration": 48.757041666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the audience listening intently on the tiered seating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the audience listening intently on the tiered seating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.6308225393295288, -0.5077633857727051], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.5898028612136841, -0.5487831234931946], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004285.pt"}
{"idx": 4287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5VmfgTfuYGE.mp4", "duration": 48.757041666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone writing on pink sticky notes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone writing on pink sticky notes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.27161121368408203, 0.39467036724090576], "gt_se_raw": [33.0, 34.0], "gt_se_norm": [0.3536505699157715, 0.39467036724090576], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004287.pt"}
{"idx": 4289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5VmfgTfuYGE.mp4", "duration": 48.757041666666666, "prompt_text": "Please find the visual event described by the sentence 'Where can I find three people having a conversation on a rooftop at sunset?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find three people having a conversation on a rooftop at sunset?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 44.0], "base_se_norm": [0.7228280305862427, 0.8048675060272217], "gt_se_raw": [42.0, 43.0], "gt_se_norm": [0.7228280305862427, 0.763847827911377], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004289.pt"}
{"idx": 4291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zCdVxICvEGg.mp4", "duration": 118.8187, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a man performs a handstand while breakdancing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a man performs a handstand while breakdancing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.32670533657073975, -0.27620822191238403], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.32670533657073975, -0.29304057359695435], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004291.pt"}
{"idx": 4293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zCdVxICvEGg.mp4", "duration": 118.8187, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a bald man is being interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a bald man is being interviewed outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 72.0], "base_se_norm": [0.04360675811767578, 0.21193039417266846], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [0.04360675811767578, 0.11093616485595703], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004293.pt"}
{"idx": 4295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V3wLiAmIrGk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when several men are gathered around a red truck with its hood open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when several men are gathered around a red truck with its hood open.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004295.pt"}
{"idx": 4297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V3wLiAmIrGk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a white truck is maneuvering in reverse between orange cones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a white truck is maneuvering in reverse between orange cones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004297.pt"}
{"idx": 4299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V3wLiAmIrGk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the text 'CDL PROGRAM' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the text 'CDL PROGRAM' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.8689835071563721, -0.5414421558380127], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.8689835071563721, -0.5414421558380127], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004299.pt"}
{"idx": 4301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cZdO2e8K29o.mp4", "duration": 190.2901, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the paper is folded diagonally to form a triangle for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the paper is folded diagonally to form a triangle for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 13.0], "base_se_norm": [-0.9369384050369263, -0.8633664846420288], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.9474486708641052, -0.8948973417282104], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004301.pt"}
{"idx": 4303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cZdO2e8K29o.mp4", "duration": 190.2901, "prompt_text": "Please find the visual event described by the sentence 'At what point is the single layer's tip folded up and over the top edge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the single layer's tip folded up and over the top edge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 84.0], "base_se_norm": [-0.21172988414764404, -0.11713749170303345], "gt_se_raw": [95.0, 102.0], "gt_se_norm": [-0.001524507999420166, 0.07204735279083252], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004303.pt"}
{"idx": 4305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F0M7n8DGjBk.mp4", "duration": 180.26341666666664, "prompt_text": "Please find the visual event described by the sentence 'When is the view of a large, striped planet visible through the clouds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the view of a large, striped planet visible through the clouds?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 26.0], "base_se_norm": [-0.9112409949302673, -0.71153324842453], "gt_se_raw": [8.0, 22.0], "gt_se_norm": [-0.9112409949302673, -0.7559127807617188], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004305.pt"}
{"idx": 4307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F0M7n8DGjBk.mp4", "duration": 180.26341666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point is an astronaut seen floating in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is an astronaut seen floating in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 145.0], "base_se_norm": [0.5865670442581177, 0.6087568998336792], "gt_se_raw": [142.0, 144.0], "gt_se_norm": [0.5754722356796265, 0.5976619720458984], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004307.pt"}
{"idx": 4309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F0M7n8DGjBk.mp4", "duration": 180.26341666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see the package of canned food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the package of canned food?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 137.0], "base_se_norm": [0.486713171005249, 0.5199978351593018], "gt_se_raw": [134.0, 136.0], "gt_se_norm": [0.486713171005249, 0.5089030265808105], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004309.pt"}
{"idx": 4311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YPFGT4ecnIU.mp4", "duration": 40.6, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the girl runs towards the camera from the living room area.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the girl runs towards the camera from the living room area.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.7536945939064026, -0.6059113144874573], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.7536945939064026, -0.6551724076271057], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004311.pt"}
{"idx": 4313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YPFGT4ecnIU.mp4", "duration": 40.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the girl runs towards the person filming, smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the girl runs towards the person filming, smiling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [0.6748769283294678, 0.8719213008880615], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.7241380214691162, 0.8226602077484131], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004313.pt"}
{"idx": 4315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w28U-Js6g1Q.mp4", "duration": 203.08, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the white tank top play the piano while smoking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the white tank top play the piano while smoking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 32.0], "base_se_norm": [-0.812881588935852, -0.6848532557487488], "gt_se_raw": [21.0, 27.0], "gt_se_norm": [-0.7931849360466003, -0.7340949773788452], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004315.pt"}
{"idx": 4317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w28U-Js6g1Q.mp4", "duration": 203.08, "prompt_text": "Please find the visual event described by the sentence 'When does the woman get angry and yell at the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman get angry and yell at the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 102.0], "base_se_norm": [-0.3204648494720459, 0.004530191421508789], "gt_se_raw": [72.0, 86.0], "gt_se_norm": [-0.29091984033584595, -0.15304315090179443], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004317.pt"}
{"idx": 4319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/w28U-Js6g1Q.mp4", "duration": 203.08, "prompt_text": "Please find the visual event described by the sentence 'When are the man and three women performing the song on stage for an audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 199.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and three women performing the song on stage for an audience?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 199.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 199.0], "base_se_norm": [0.605278730392456, 0.9598187208175659], "gt_se_raw": [163.0, 170.0], "gt_se_norm": [0.605278730392456, 0.6742169857025146], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004319.pt"}
{"idx": 4321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K23OHlnu0e0.mp4", "duration": 65.99926666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman hold the object with both of her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman hold the object with both of her hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.39393264055252075, -0.151505708694458], "gt_se_raw": [20.0, 28.0], "gt_se_norm": [-0.39393264055252075, -0.151505708694458], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004321.pt"}
{"idx": 4323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K23OHlnu0e0.mp4", "duration": 65.99926666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the speaker mentions the 'sensory systems'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the speaker mentions the 'sensory systems'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.09089899063110352, 0.03031444549560547], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [0.2121347188949585, 0.272741436958313], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004323.pt"}
{"idx": 4325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0GtT-vN33I.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the white shirt sanding the side of a vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the white shirt sanding the side of a vehicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5409404635429382], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004325.pt"}
{"idx": 4327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0GtT-vN33I.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an older man in a red shirt is polishing a car hood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an older man in a red shirt is polishing a car hood.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.016300976276397705, 0.3771786689758301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004327.pt"}
{"idx": 4329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0GtT-vN33I.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the text 'LENGTH of PROGRAM' appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the text 'LENGTH of PROGRAM' appears on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.049278974533081055, 0.3771786689758301], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004329.pt"}
{"idx": 4331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzzjgBAaWZw.mp4", "duration": 73.2732, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat peek from behind the black furniture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat peek from behind the black furniture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 68.0], "base_se_norm": [0.5012309551239014, 0.856067419052124], "gt_se_raw": [55.0, 61.0], "gt_se_norm": [0.5012309551239014, 0.665001630783081], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004331.pt"}
{"idx": 4333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fzzjgBAaWZw.mp4", "duration": 73.2732, "prompt_text": "Please find the visual event described by the sentence 'Locate the exact time the cat lunges at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the exact time the cat lunges at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 69.0], "base_se_norm": [0.7741820812225342, 0.8833625316619873], "gt_se_raw": [67.0, 69.0], "gt_se_norm": [0.8287723064422607, 0.8833625316619873], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004333.pt"}
{"idx": 4335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahSKGi_21xg.mp4", "duration": 163.56356356356358, "prompt_text": "Please find the visual event described by the sentence 'When is the man with short hair and a brown jacket reading from his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man with short hair and a brown jacket reading from his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.9388616681098938, -0.853268027305603], "gt_se_raw": [5.0, 12.0], "gt_se_norm": [-0.9388616681098938, -0.853268027305603], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004335.pt"}
{"idx": 4337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahSKGi_21xg.mp4", "duration": 163.56356356356358, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the man in the blue knitted cardigan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the man in the blue knitted cardigan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 54.0], "base_se_norm": [-0.41307222843170166, -0.3397062420845032], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [-0.4008445739746094, -0.3397062420845032], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004337.pt"}
{"idx": 4339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ahSKGi_21xg.mp4", "duration": 163.56356356356358, "prompt_text": "Please find the visual event described by the sentence 'When are the three men, one with green hair, on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three men, one with green hair, on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 151.0], "base_se_norm": [0.6507343053817749, 0.846376895904541], "gt_se_raw": [135.0, 152.0], "gt_se_norm": [0.6507343053817749, 0.8586045503616333], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004339.pt"}
{"idx": 4341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZFpnzOGnZBM.mp4", "duration": 128.39493333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the video introducing the Grit Freedom Chair product?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the video introducing the Grit Freedom Chair product?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 52.0], "base_se_norm": [-0.4548071026802063, -0.18999916315078735], "gt_se_raw": [34.0, 39.0], "gt_se_norm": [-0.47038406133651733, -0.3924993872642517], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004341.pt"}
{"idx": 4343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZFpnzOGnZBM.mp4", "duration": 128.39493333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the three models of the wheelchair, 3.0, Spartan, and Pro, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the three models of the wheelchair, 3.0, Spartan, and Pro, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 59.0], "base_se_norm": [-0.18999916315078735, -0.08096057176589966], "gt_se_raw": [52.0, 58.0], "gt_se_norm": [-0.18999916315078735, -0.0965375304222107], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004343.pt"}
{"idx": 4345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3aZadizk_Io.mp4", "duration": 162.12863333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the monkey, Rafiki, guiding the lion through a dense, thorny thicket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the monkey, Rafiki, guiding the lion through a dense, thorny thicket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 38.0], "base_se_norm": [-0.9876641035079956, -0.5312364101409912], "gt_se_raw": [10.0, 38.0], "gt_se_norm": [-0.8766411542892456, -0.5312364101409912], "start_token_span": [76, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004345.pt"}
{"idx": 4347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3aZadizk_Io.mp4", "duration": 162.12863333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the ghost of a lion materializes in the clouds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the ghost of a lion materializes in the clouds.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 110.0], "base_se_norm": [0.23358845710754395, 0.35694730281829834], "gt_se_raw": [100.0, 109.0], "gt_se_norm": [0.23358845710754395, 0.34461140632629395], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004347.pt"}
{"idx": 4349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3aZadizk_Io.mp4", "duration": 162.12863333333334, "prompt_text": "Please find the visual event described by the sentence 'During which period does the spirit in the clouds speak to the lion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'During which period does the spirit in the clouds speak to the lion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 127.0], "base_se_norm": [0.2705960273742676, 0.566657304763794], "gt_se_raw": [109.0, 142.0], "gt_se_norm": [0.34461140632629395, 0.7516955137252808], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004349.pt"}
{"idx": 4351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oyv6hjQr8MA.mp4", "duration": 124.75796666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the time range for the detailed explanation of the Plus Meal Plan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range for the detailed explanation of the Plus Meal Plan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 41.0], "base_se_norm": [-0.5350998640060425, -0.3427273631095886], "gt_se_raw": [33.0, 41.0], "gt_se_norm": [-0.4709756374359131, -0.3427273631095886], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004351.pt"}
{"idx": 4353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oyv6hjQr8MA.mp4", "duration": 124.75796666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video provide a visual example of items that can be purchased with I-BUCKS?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video provide a visual example of items that can be purchased with I-BUCKS?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 102.0], "base_se_norm": [0.5710419416427612, 0.6351661682128906], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [0.5710419416427612, 0.6351661682128906], "start_token_span": [76, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004353.pt"}
{"idx": 4355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sBdqOWSZ56w.mp4", "duration": 49.3, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of creating the first stitch on the needle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of creating the first stitch on the needle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 36.0], "base_se_norm": [-0.1480730175971985, 0.46044623851776123], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.10750508308410645, 0.13590264320373535], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004355.pt"}
{"idx": 4357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sBdqOWSZ56w.mp4", "duration": 49.3, "prompt_text": "Please find the visual event described by the sentence 'When can we observe multiple stitches being added to the needle in repetition?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe multiple stitches being added to the needle in repetition?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 37.0], "base_se_norm": [-0.10750508308410645, 0.501014232635498], "gt_se_raw": [29.0, 36.0], "gt_se_norm": [0.17647063732147217, 0.46044623851776123], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004357.pt"}
{"idx": 4359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sBdqOWSZ56w.mp4", "duration": 49.3, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the needle picks up the loop around the index finger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the needle picks up the loop around the index finger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 36.0], "base_se_norm": [-0.1480730175971985, 0.46044623851776123], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [0.014198780059814453, 0.09533464908599854], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004359.pt"}
{"idx": 4361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ineZXLbL7s8.mp4", "duration": 100.0, "prompt_text": "Please find the visual event described by the sentence 'At what moment is a person in a yellow hazmat suit using a fire extinguisher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is a person in a yellow hazmat suit using a fire extinguisher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 80.0], "base_se_norm": [0.440000057220459, 0.6000000238418579], "gt_se_raw": [71.0, 80.0], "gt_se_norm": [0.41999995708465576, 0.6000000238418579], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004361.pt"}
{"idx": 4363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ineZXLbL7s8.mp4", "duration": 100.0, "prompt_text": "Please find the visual event described by the sentence 'When is the photographer using a phone to create a reflection?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the photographer using a phone to create a reflection?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.8999999761581421, -0.8400000333786011], "gt_se_raw": [5.0, 12.0], "gt_se_norm": [-0.8999999761581421, -0.7599999904632568], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004363.pt"}
{"idx": 4365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zMqQNCmFQE.mp4", "duration": 100.16683350016683, "prompt_text": "Please find the visual event described by the sentence 'When is the can being heated by a blowtorch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the can being heated by a blowtorch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 35.0], "base_se_norm": [-0.8003331422805786, -0.3011658787727356], "gt_se_raw": [10.0, 35.0], "gt_se_norm": [-0.8003331422805786, -0.3011658787727356], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004365.pt"}
{"idx": 4367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zMqQNCmFQE.mp4", "duration": 100.16683350016683, "prompt_text": "Please find the visual event described by the sentence 'At what point does a hand use the can to punch a hole through the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a hand use the can to punch a hole through the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 64.0], "base_se_norm": [0.19800138473510742, 0.2778681516647339], "gt_se_raw": [55.0, 65.0], "gt_se_norm": [0.09816789627075195, 0.29783475399017334], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004367.pt"}
{"idx": 4369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zMqQNCmFQE.mp4", "duration": 100.16683350016683, "prompt_text": "Please find the visual event described by the sentence 'When are beans poured onto a person's back and made into a sandwich?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are beans poured onto a person's back and made into a sandwich?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 93.0], "base_se_norm": [0.6971685886383057, 0.8569021224975586], "gt_se_raw": [86.0, 93.0], "gt_se_norm": [0.7171353101730347, 0.8569021224975586], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004369.pt"}
{"idx": 4371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CIJjyEK9jB8.mp4", "duration": 189.95643333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot of the black dog lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot of the black dog lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.8841840028762817, -0.8631265163421631], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.8736552596092224, -0.8631265163421631], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004371.pt"}
{"idx": 4373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CIJjyEK9jB8.mp4", "duration": 189.95643333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a woman is using an iPad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a woman is using an iPad.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.9578850865364075, -0.9262989163398743], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9578850865364075, -0.9368276000022888], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004373.pt"}
{"idx": 4375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1a7wwE952_E.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'When is a person using a tool to work on the engine of a white car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person using a tool to work on the engine of a white car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.540437638759613], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.540437638759613], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004375.pt"}
{"idx": 4377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1a7wwE952_E.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a woman with purple hair is repairing a car from below.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a woman with purple hair is repairing a car from below.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4443387985229492, 0.8382494449615479], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4443387985229492, 0.8382494449615479], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004377.pt"}
{"idx": 4379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1a7wwE952_E.mp4", "duration": 30.463766666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man in a red shirt is pointing at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man in a red shirt is pointing at a computer screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.015223503112792969, 0.3786870241165161], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.015223503112792969, 0.3786870241165161], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004379.pt"}
{"idx": 4381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-7-quPaRol8.mp4", "duration": 187.12026666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a quote from Dr. Caroline Solomon is displayed on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a quote from Dr. Caroline Solomon is displayed on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 54.0], "base_se_norm": [-0.4762726426124573, -0.42283111810684204], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [-0.4762726426124573, -0.412142813205719], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004381.pt"}
{"idx": 4383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-7-quPaRol8.mp4", "duration": 187.12026666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the video shows a picture of Susanna Harris.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the video shows a picture of Susanna Harris.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 119.0], "base_se_norm": [0.2505326271057129, 0.271909236907959], "gt_se_raw": [117.0, 120.0], "gt_se_norm": [0.2505326271057129, 0.28259754180908203], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004383.pt"}
{"idx": 4385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aKIXMD3NQIA.mp4", "duration": 182.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the adult holding the duckling and interacting with the child?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the adult holding the duckling and interacting with the child?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 34.0], "base_se_norm": [-0.934402346611023, -0.6282799243927002], "gt_se_raw": [7.0, 33.0], "gt_se_norm": [-0.9234693646430969, -0.6392128467559814], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004385.pt"}
{"idx": 4387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aKIXMD3NQIA.mp4", "duration": 182.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what time is the woman using a comb on the girl's hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the woman using a comb on the girl's hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 122.0], "base_se_norm": [-0.23469388484954834, 0.33381927013397217], "gt_se_raw": [83.0, 87.0], "gt_se_norm": [-0.09256559610366821, -0.04883384704589844], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004387.pt"}
{"idx": 4389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aKIXMD3NQIA.mp4", "duration": 182.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the student is drawing on a piece of paper with a pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the student is drawing on a piece of paper with a pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 182.0], "base_se_norm": [0.9569970369338989, 0.9897959232330322], "gt_se_raw": [179.0, 182.0], "gt_se_norm": [0.9569970369338989, 0.9897959232330322], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004389.pt"}
{"idx": 4391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nwB2P3g1upM.mp4", "duration": 484.6, "prompt_text": "Please find the visual event described by the sentence 'When are the beads cleared to reset the abacus after the first problem is solved?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the beads cleared to reset the abacus after the first problem is solved?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 115.0], "base_se_norm": [-0.5625258088111877, -0.5253818035125732], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.7977713346481323, -0.7895171642303467], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004391.pt"}
{"idx": 4393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nwB2P3g1upM.mp4", "duration": 484.6, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the abacus shows the result 20.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the abacus shows the result 20.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 50.0], "base_se_norm": [-0.8184069395065308, -0.7936442494392395], "gt_se_raw": [43.0, 48.0], "gt_se_norm": [-0.8225340843200684, -0.8018984794616699], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004393.pt"}
{"idx": 4395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oPwrodxghrw.mp4", "duration": 8.875533333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the dark suit inspecting a bill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the dark suit inspecting a bill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.3239842653274536], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.3239842653274536], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004395.pt"}
{"idx": 4397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oPwrodxghrw.mp4", "duration": 8.875533333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the grey suit make a gesture indicating a small amount?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the grey suit make a gesture indicating a small amount?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.09864568710327148, 0.5773700475692749], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.3239842653274536, 0.12669289112091064], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004397.pt"}
{"idx": 4399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oPwrodxghrw.mp4", "duration": 8.875533333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when one man is holding two bags while the other makes a hand gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when one man is holding two bags while the other makes a hand gesture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, 0.802708625793457], "gt_se_raw": [3.0, 8.0], "gt_se_norm": [-0.3239842653274536, 0.802708625793457], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004399.pt"}
{"idx": 4401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ycvwnsZ3ZCo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When is a person using a pipette to transfer red liquid into a petri dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person using a pipette to transfer red liquid into a petri dish?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.4753605127334595, -0.08188092708587646], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004401.pt"}
{"idx": 4403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ycvwnsZ3ZCo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the 'LENGTH of PROGRAM' title appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the 'LENGTH of PROGRAM' title appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.049278974533081055, 0.3771786689758301], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004403.pt"}
{"idx": 4405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rj07zM_GLeY.mp4", "duration": 51.38466666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman playfully move the wipe towards the boy's shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman playfully move the wipe towards the boy's shoe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.22155767679214478, 0.08981931209564209], "gt_se_raw": [20.0, 24.0], "gt_se_norm": [-0.22155767679214478, -0.06586921215057373], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004405.pt"}
{"idx": 4407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rj07zM_GLeY.mp4", "duration": 51.38466666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman laughs and playfully moves the wipe after the boy laughs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman laughs and playfully moves the wipe after the boy laughs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 51.0], "base_se_norm": [0.8682615756988525, 0.9850279092788696], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [0.4790405035018921, 0.5958068370819092], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004407.pt"}
{"idx": 4409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Rj07zM_GLeY.mp4", "duration": 51.38466666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the two people passing the wipe back and forth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two people passing the wipe back and forth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 49.0], "base_se_norm": [0.24550771713256836, 0.9071837663650513], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [0.08981931209564209, 0.24550771713256836], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004409.pt"}
{"idx": 4411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gPdgsPTNhac.mp4", "duration": 141.57476666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man with glasses is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man with glasses is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 38.0], "base_se_norm": [-0.5338152647018433, -0.46318119764328003], "gt_se_raw": [32.0, 38.0], "gt_se_norm": [-0.547942042350769, -0.46318119764328003], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004411.pt"}
{"idx": 4413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gPdgsPTNhac.mp4", "duration": 141.57476666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip showing geysers erupting from a moon's surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip showing geysers erupting from a moon's surface.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 61.0], "base_se_norm": [-0.2088986039161682, -0.13826453685760498], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [-0.2088986039161682, -0.15239131450653076], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004413.pt"}
{"idx": 4415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txy3Joh7DZ4.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the instructor use her index finger to demonstrate moving up the drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the instructor use her index finger to demonstrate moving up the drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 57.0], "base_se_norm": [0.25553321838378906, 0.3762575387954712], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.758551299571991, -0.6861166954040527], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004415.pt"}
{"idx": 4417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txy3Joh7DZ4.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the instructor drawing the black arrow on the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the instructor drawing the black arrow on the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [-0.010060369968414307, 0.08651912212371826], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.010060369968414307, 0.062374234199523926], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004417.pt"}
{"idx": 4419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/txy3Joh7DZ4.mp4", "duration": 82.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the instructor hold up the paper to show the final drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the instructor hold up the paper to show the final drawing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 72.0], "base_se_norm": [0.5211267471313477, 0.7384305000305176], "gt_se_raw": [64.0, 68.0], "gt_se_norm": [0.545271635055542, 0.6418510675430298], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004419.pt"}
{"idx": 4421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G64wuf-rHoo.mp4", "duration": 46.046, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a green t-shirt is presenting in front of an NFL logo wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a green t-shirt is presenting in front of an NFL logo wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 7.0], "base_se_norm": [-0.7828258872032166, -0.6959562301635742], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [-0.7828258872032166, -0.6959562301635742], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004421.pt"}
{"idx": 4423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G64wuf-rHoo.mp4", "duration": 46.046, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where several people are writing at a green table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where several people are writing at a green table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.174738347530365, -0.0444338321685791], "gt_se_raw": [18.0, 22.0], "gt_se_norm": [-0.21817314624786377, -0.0444338321685791], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004423.pt"}
{"idx": 4425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3jb5aI4ejo.mp4", "duration": 298.3814111111111, "prompt_text": "Please find the visual event described by the sentence 'When does the woman make her bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman make her bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 43.0], "base_se_norm": [-0.8257263898849487, -0.7117782831192017], "gt_se_raw": [25.0, 43.0], "gt_se_norm": [-0.8324292302131653, -0.7117782831192017], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004425.pt"}
{"idx": 4427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3jb5aI4ejo.mp4", "duration": 298.3814111111111, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman running outside while listening to music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman running outside while listening to music?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 119.0], "base_se_norm": [-0.32301414012908936, -0.20236319303512573], "gt_se_raw": [98.0, 119.0], "gt_se_norm": [-0.34312260150909424, -0.20236319303512573], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004427.pt"}
{"idx": 4429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P3jb5aI4ejo.mp4", "duration": 298.3814111111111, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where she is tidying up her messy room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 194.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where she is tidying up her messy room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 194.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 194.0], "base_se_norm": [0.13277840614318848, 0.3003491163253784], "gt_se_raw": [170.0, 194.0], "gt_se_norm": [0.13948118686676025, 0.3003491163253784], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004429.pt"}
{"idx": 4431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/khmmhvlCUNQ.mp4", "duration": 186.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the website's central slideshow displays an image of jewelry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the website's central slideshow displays an image of jewelry.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 24.0], "base_se_norm": [-0.8285101652145386, -0.7427653074264526], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [-0.8392282724380493, -0.8070739507675171], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004431.pt"}
{"idx": 4433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/khmmhvlCUNQ.mp4", "duration": 186.6, "prompt_text": "Please find the visual event described by the sentence 'When is the footer section of the webpage, containing links like 'Sitemap' and 'Privacy Policy', being discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the footer section of the webpage, containing links like 'Sitemap' and 'Privacy Policy', being discussed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 132.0], "base_se_norm": [0.3183279037475586, 0.4147909879684448], "gt_se_raw": [136.0, 148.0], "gt_se_norm": [0.4576634168624878, 0.5862807035446167], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004433.pt"}
{"idx": 4435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KItqbZXlrdY.mp4", "duration": 99.84974444444444, "prompt_text": "Please find the visual event described by the sentence 'When are the potatoes being mashed in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the potatoes being mashed in the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.7596388459205627, -0.679518461227417], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.7396087646484375, -0.6594883799552917], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004435.pt"}
{"idx": 4437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KItqbZXlrdY.mp4", "duration": 99.84974444444444, "prompt_text": "Please find the visual event described by the sentence 'At what point are the croquettes being shaped and placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the croquettes being shaped and placed on the baking sheet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 42.0], "base_se_norm": [-0.2989466190338135, -0.158735990524292], "gt_se_raw": [34.0, 42.0], "gt_se_norm": [-0.3189767599105835, -0.158735990524292], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004437.pt"}
{"idx": 4439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KItqbZXlrdY.mp4", "duration": 99.84974444444444, "prompt_text": "Please find the visual event described by the sentence 'When are the croquettes being deep-fried in the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the croquettes being deep-fried in the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 64.0], "base_se_norm": [0.141715407371521, 0.28192615509033203], "gt_se_raw": [58.0, 63.0], "gt_se_norm": [0.16174554824829102, 0.261896014213562], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004439.pt"}
{"idx": 4441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UlgYIQrcYnQ.mp4", "duration": 25.442083333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the Venus flytrap is being fed the pickle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the Venus flytrap is being fed the pickle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.056681036949157715, 0.25775861740112305], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.021928906440734863, 0.17914867401123047], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004441.pt"}
{"idx": 4443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UlgYIQrcYnQ.mp4", "duration": 25.442083333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a potted plant with the name 'CHARLIE' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a potted plant with the name 'CHARLIE' is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.7641702890396118, -0.6069504022598267], "gt_se_raw": [3.0, 4.0], "gt_se_norm": [-0.7641702890396118, -0.6855603456497192], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004443.pt"}
{"idx": 4445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0rrS51ry2s.mp4", "duration": 112.82103333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the parents pointing their fingers towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the parents pointing their fingers towards the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 25.0], "base_se_norm": [-0.6454561948776245, -0.5568202137947083], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.6454561948776245, -0.5922746062278748], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004445.pt"}
{"idx": 4447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0rrS51ry2s.mp4", "duration": 112.82103333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man is playing peek-a-boo from behind the green sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man is playing peek-a-boo from behind the green sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 49.0], "base_se_norm": [-0.22000354528427124, -0.13136762380599976], "gt_se_raw": [46.0, 49.0], "gt_se_norm": [-0.1845492124557495, -0.13136762380599976], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004447.pt"}
{"idx": 4449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0rrS51ry2s.mp4", "duration": 112.82103333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the baby playing with a colorful bead maze toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the baby playing with a colorful bead maze toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 67.0], "base_se_norm": [0.09908592700958252, 0.187721848487854], "gt_se_raw": [62.0, 67.0], "gt_se_norm": [0.09908592700958252, 0.187721848487854], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004449.pt"}
{"idx": 4451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BnZtKCcMxzE.mp4", "duration": 128.9288, "prompt_text": "Please find the visual event described by the sentence 'At what point does Ellen hold up the black smart speaker and begin whispering to it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Ellen hold up the black smart speaker and begin whispering to it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 63.0], "base_se_norm": [-0.22437810897827148, -0.0227164626121521], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.16232836246490479, -0.11579108238220215], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004451.pt"}
{"idx": 4453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BnZtKCcMxzE.mp4", "duration": 128.9288, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man in the hat laughing and covering his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man in the hat laughing and covering his face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 122.0], "base_se_norm": [0.8614925146102905, 0.8925173282623291], "gt_se_raw": [120.0, 122.0], "gt_se_norm": [0.8614925146102905, 0.8925173282623291], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004453.pt"}
{"idx": 4455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_PklVas9cA.mp4", "duration": 141.161, "prompt_text": "Please find the visual event described by the sentence 'When does Ross show the others a sketch of his plan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Ross show the others a sketch of his plan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 50.0], "base_se_norm": [-0.7024673819541931, -0.29158902168273926], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.7024673819541931, -0.6316262483596802], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004455.pt"}
{"idx": 4457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_PklVas9cA.mp4", "duration": 141.161, "prompt_text": "Please find the visual event described by the sentence 'At what point does Ross repeatedly shout 'Pivot'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Ross repeatedly shout 'Pivot'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 54.0], "base_se_norm": [-0.44743943214416504, -0.23491615056991577], "gt_se_raw": [82.0, 94.0], "gt_se_norm": [0.16179406642913818, 0.33181262016296387], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004457.pt"}
{"idx": 4459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_PklVas9cA.mp4", "duration": 141.161, "prompt_text": "Please find the visual event described by the sentence 'When is the couch visibly stuck in the stairwell, and they stop trying to move it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 127.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the couch visibly stuck in the stairwell, and they stop trying to move it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 127.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 127.0], "base_se_norm": [0.5726723670959473, 0.7993638515472412], "gt_se_raw": [97.0, 108.0], "gt_se_norm": [0.3743172883987427, 0.5301676988601685], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004459.pt"}
{"idx": 4461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ixh8KqEr6LE.mp4", "duration": 115.03158333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the bottle is being filled with water from the tap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the bottle is being filled with water from the tap.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 63.0], "base_se_norm": [-0.09590047597885132, 0.09535133838653564], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [-0.02635437250137329, 0.04319179058074951], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004461.pt"}
{"idx": 4463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ixh8KqEr6LE.mp4", "duration": 115.03158333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman puts the mop pad into the washing machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman puts the mop pad into the washing machine.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 103.0], "base_se_norm": [0.6864932775497437, 0.7908124923706055], "gt_se_raw": [99.0, 101.0], "gt_se_norm": [0.7212663888931274, 0.7560393810272217], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004463.pt"}
{"idx": 4465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HoKkasEyDOI.mp4", "duration": 195.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the line graph comparing US murder and robbery rates per 100,000.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the line graph comparing US murder and robbery rates per 100,000.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 15.0], "base_se_norm": [-0.8673187494277954, -0.8469062447547913], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.8673187494277954, -0.8469062447547913], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004465.pt"}
{"idx": 4467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HoKkasEyDOI.mp4", "duration": 195.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the bar chart showing arrest rates for schools with and without school resource officers appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bar chart showing arrest rates for schools with and without school resource officers appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [-0.4182436466217041, -0.3672124147415161], "gt_se_raw": [57.0, 62.0], "gt_se_norm": [-0.4182436466217041, -0.3672124147415161], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004467.pt"}
{"idx": 4469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HoKkasEyDOI.mp4", "duration": 195.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When are weapons, drugs, alcohol, and assault depicted as cartoon cutouts on sticks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are weapons, drugs, alcohol, and assault depicted as cartoon cutouts on sticks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 94.0], "base_se_norm": [-0.18349987268447876, -0.04061233997344971], "gt_se_raw": [90.0, 93.0], "gt_se_norm": [-0.08143734931945801, -0.05081862211227417], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004469.pt"}
{"idx": 4471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9MO1aY1xC80.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'At what point does a leaf fall and break the line of ants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a leaf fall and break the line of ants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.25074928998947144, -0.20079922676086426], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.4255744218826294, -0.3256743550300598], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004471.pt"}
{"idx": 4473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9MO1aY1xC80.mp4", "duration": 80.08, "prompt_text": "Please find the visual event described by the sentence 'When does the lead ant instruct the others to go around the leaf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the lead ant instruct the others to go around the leaf?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 53.0], "base_se_norm": [0.048951029777526855, 0.32367634773254395], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.0009990334510803223, 0.07392609119415283], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004473.pt"}
{"idx": 4475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/65WL8FfN-Io.mp4", "duration": 174.17399999999998, "prompt_text": "Please find the visual event described by the sentence 'At what moment does a man breathe fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does a man breathe fire?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 2.0], "base_se_norm": [-0.9885172247886658, -0.9770344495773315], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.9770344495773315, -0.9655516743659973], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004475.pt"}
{"idx": 4477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/65WL8FfN-Io.mp4", "duration": 174.17399999999998, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the couple kisses passionately in front of the graffiti wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the couple kisses passionately in front of the graffiti wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 99.0], "base_se_norm": [0.06789767742156982, 0.13679420948028564], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.11382877826690674, 0.13679420948028564], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004477.pt"}
{"idx": 4479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/65WL8FfN-Io.mp4", "duration": 174.17399999999998, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for the aerial silk performance on a hoop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for the aerial silk performance on a hoop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 2.0], "base_se_norm": [-0.9885172247886658, -0.9770344495773315], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9885172247886658, -0.9770344495773315], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004479.pt"}
{"idx": 4481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v_dbomeVN2c.mp4", "duration": 29.904874999999997, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a green shirt shown downloading the app?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a green shirt shown downloading the app?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.3980914354324341, -0.26433396339416504], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.4649701714515686, -0.26433396339416504], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004481.pt"}
{"idx": 4483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v_dbomeVN2c.mp4", "duration": 29.904874999999997, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a tablet is used to scan the tiger logo on a wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a tablet is used to scan the tiger logo on a wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 20.0], "base_se_norm": [0.2706958055496216, 0.3375746011734009], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [0.2706958055496216, 0.4044532775878906], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004483.pt"}
{"idx": 4485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGnIEvFtSCY.mp4", "duration": 147.71423333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the white shirt guide the other person's hand to the keyboard symbol?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the white shirt guide the other person's hand to the keyboard symbol?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9864603281021118, -0.8646034002304077], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.8646034002304077, -0.8239844441413879], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004485.pt"}
{"idx": 4487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGnIEvFtSCY.mp4", "duration": 147.71423333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where the instructor assists in signing for 'swimming'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where the instructor assists in signing for 'swimming'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 17.0], "base_se_norm": [-0.8646034002304077, -0.76982581615448], "gt_se_raw": [51.0, 56.0], "gt_se_norm": [-0.3094775080680847, -0.24177920818328857], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004487.pt"}
{"idx": 4489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mGnIEvFtSCY.mp4", "duration": 147.71423333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the student reach for the cracker symbol at the end of the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the student reach for the cracker symbol at the end of the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.7427465319633484, -0.6479688882827759], "gt_se_raw": [97.0, 101.0], "gt_se_norm": [0.3133467435836792, 0.3675053119659424], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004489.pt"}
{"idx": 4491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pYekAIt9wW8.mp4", "duration": 39.63333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the woman in green points to the lower section of the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the woman in green points to the lower section of the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 38.0], "base_se_norm": [0.16063916683197021, 0.9175777435302734], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.16063916683197021, 0.2615642547607422], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004491.pt"}
{"idx": 4493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pYekAIt9wW8.mp4", "duration": 39.63333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where they are touching the calendar cards while singing the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where they are touching the calendar cards while singing the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 38.0], "base_se_norm": [0.21110177040100098, 0.9175777435302734], "gt_se_raw": [29.0, 38.0], "gt_se_norm": [0.4634145498275757, 0.9175777435302734], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004493.pt"}
{"idx": 4495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dSf32CWJYBs.mp4", "duration": 110.735625, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the straw hat introduce himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the straw hat introduce himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 23.0], "base_se_norm": [-0.7471455335617065, -0.5845961570739746], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.7652065753936768, -0.6929624080657959], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004495.pt"}
{"idx": 4497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dSf32CWJYBs.mp4", "duration": 110.735625, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man in the purple shirt starts praying for the disabled people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man in the purple shirt starts praying for the disabled people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 66.0], "base_se_norm": [0.029479026794433594, 0.19202828407287598], "gt_se_raw": [51.0, 60.0], "gt_se_norm": [-0.0788872241973877, 0.08366203308105469], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004497.pt"}
{"idx": 4499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dSf32CWJYBs.mp4", "duration": 110.735625, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the text 'DON'T DO THIS' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the text 'DON'T DO THIS' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 43.0], "base_se_norm": [-0.2956196665763855, -0.22337549924850464], "gt_se_raw": [40.0, 41.0], "gt_se_norm": [-0.2775586247444153, -0.25949758291244507], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004499.pt"}
{"idx": 4501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A1T3dAvFeio.mp4", "duration": 165.0, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the yellow bird runs out the door holding an egg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the yellow bird runs out the door holding an egg.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 136.0], "base_se_norm": [0.6000000238418579, 0.6484848260879517], "gt_se_raw": [133.0, 136.0], "gt_se_norm": [0.6121212244033813, 0.6484848260879517], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004501.pt"}
{"idx": 4503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A1T3dAvFeio.mp4", "duration": 165.0, "prompt_text": "Please find the visual event described by the sentence 'At what point are the three pigs locked inside the cage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 136.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the three pigs locked inside the cage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 136.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [136.0, 143.0], "base_se_norm": [0.6484848260879517, 0.7333333492279053], "gt_se_raw": [137.0, 139.0], "gt_se_norm": [0.6606060266494751, 0.684848427772522], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004503.pt"}
{"idx": 4505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebqxuyvp1qk.mp4", "duration": 39.005633333333336, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in red presenting a teddy bear to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in red presenting a teddy bear to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 28.0], "base_se_norm": [-0.12833106517791748, 0.4356900453567505], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.12833106517791748, 0.17931675910949707], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004505.pt"}
{"idx": 4507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebqxuyvp1qk.mp4", "duration": 39.005633333333336, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman brings the large book to the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman brings the large book to the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.589513897895813, 0.743337869644165], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [0.589513897895813, 0.743337869644165], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004507.pt"}
{"idx": 4509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebqxuyvp1qk.mp4", "duration": 39.005633333333336, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'Clip #1: Liz Introduces The Book To Nia' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'Clip #1: Liz Introduces The Book To Nia' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 14.0], "base_se_norm": [-0.5385282039642334, -0.28215497732162476], "gt_se_raw": [9.0, 14.0], "gt_se_norm": [-0.5385282039642334, -0.28215497732162476], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004509.pt"}
{"idx": 4511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JnR2dpLnS14.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a man's feet in crampons are seen crossing a ladder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a man's feet in crampons are seen crossing a ladder.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 88.0], "base_se_norm": [-0.10890769958496094, -0.007390797138214111], "gt_se_raw": [79.0, 88.0], "gt_se_norm": [-0.10890769958496094, -0.007390797138214111], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004511.pt"}
{"idx": 4513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JnR2dpLnS14.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'When is a man speaking inside a tent, his face lit by a red light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a man speaking inside a tent, his face lit by a red light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 139.0], "base_se_norm": [0.364837646484375, 0.5678713321685791], "gt_se_raw": [121.0, 139.0], "gt_se_norm": [0.364837646484375, 0.5678713321685791], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004513.pt"}
{"idx": 4515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J0HiXwK5s2k.mp4", "duration": 63.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the sign for the Human Performance Clinical/Research Laboratory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the sign for the Human Performance Clinical/Research Laboratory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.8730158805847168], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9047619104385376], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004515.pt"}
{"idx": 4517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J0HiXwK5s2k.mp4", "duration": 63.0, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a person in blue gloves operates a computer mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a person in blue gloves operates a computer mouse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [-0.4603174328804016, -0.39682537317276], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.4603174328804016, -0.39682537317276], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004517.pt"}
{"idx": 4519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/J0HiXwK5s2k.mp4", "duration": 63.0, "prompt_text": "Please find the visual event described by the sentence 'At what time are two researchers seen collaborating at the lab station?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are two researchers seen collaborating at the lab station?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 52.0], "base_se_norm": [0.3968254327774048, 0.6507936716079712], "gt_se_raw": [43.0, 52.0], "gt_se_norm": [0.365079402923584, 0.6507936716079712], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004519.pt"}
{"idx": 4521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsSq3xiTWC8.mp4", "duration": 87.4874, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the girl smells the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the girl smells the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 42.0], "base_se_norm": [-0.3599078059196472, -0.03986167907714844], "gt_se_raw": [28.0, 36.0], "gt_se_norm": [-0.3599078059196472, -0.17702430486679077], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004521.pt"}
{"idx": 4523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsSq3xiTWC8.mp4", "duration": 87.4874, "prompt_text": "Please find the visual event described by the sentence 'Locate the interval where the woman is kneading the dough with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the interval where the woman is kneading the dough with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.4742099642753601], "gt_se_raw": [0.0, 23.0], "gt_se_norm": [-1.0, -0.4742099642753601], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004523.pt"}
{"idx": 4525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I2yW_VONWco.mp4", "duration": 406.0, "prompt_text": "Please find the visual event described by the sentence 'When are the egg yolks added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the egg yolks added to the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [-0.7586206793785095, -0.7241379022598267], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [-0.7586206793785095, -0.7290640473365784], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004525.pt"}
{"idx": 4527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I2yW_VONWco.mp4", "duration": 406.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the red dough rolled into a rectangular shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the red dough rolled into a rectangular shape?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 164.0], "base_se_norm": [-0.2660098671913147, -0.19211822748184204], "gt_se_raw": [148.0, 164.0], "gt_se_norm": [-0.27093595266342163, -0.19211822748184204], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004527.pt"}
{"idx": 4529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I2yW_VONWco.mp4", "duration": 406.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the rainbow-colored dough log is being sliced to reveal the heart shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 283.0 - 304.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the rainbow-colored dough log is being sliced to reveal the heart shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 283.0 - 304.0 seconds.<|im_end|>\n", "base_se_raw": [283.0, 304.0], "base_se_norm": [0.39408862590789795, 0.49753689765930176], "gt_se_raw": [284.0, 288.0], "gt_se_norm": [0.39901483058929443, 0.41871917247772217], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004529.pt"}
{"idx": 4531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aJV1WVWbyNU.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a cat is stretching up against the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a cat is stretching up against the door.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.8082459568977356, -0.7405680418014526], "gt_se_raw": [18.0, 23.0], "gt_se_norm": [-0.7969663143157959, -0.7405680418014526], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004531.pt"}
{"idx": 4533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aJV1WVWbyNU.mp4", "duration": 177.31046666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where one of the cats is rolling on its back on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where one of the cats is rolling on its back on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 126.0], "base_se_norm": [0.3197190761566162, 0.42123591899871826], "gt_se_raw": [119.0, 125.0], "gt_se_norm": [0.3422783613204956, 0.4099562168121338], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004533.pt"}
{"idx": 4535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6lZBoR5gvs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the narrator first compare the employment rates of disabled and non-disabled people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the narrator first compare the employment rates of disabled and non-disabled people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.03166300058364868], "gt_se_raw": [1.0, 6.0], "gt_se_norm": [-0.9031662940979004, -0.4189978241920471], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004535.pt"}
{"idx": 4537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6lZBoR5gvs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is a general trend regarding the employment of people with disabilities mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a general trend regarding the employment of people with disabilities mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9031662940979004, -0.03166300058364868], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.2253304123878479, 0.355671763420105], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004537.pt"}
{"idx": 4539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6lZBoR5gvs.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'What is the full duration of the speaker's explanation of the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the full duration of the speaker's explanation of the chart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [0.0, 14.0], "gt_se_norm": [-1.0, 0.355671763420105], "start_token_span": [74, 77], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004539.pt"}
{"idx": 4541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S1Cn8vIGlLo.mp4", "duration": 66.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when the baby appears to be crying or upset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when the baby appears to be crying or upset.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 34.0], "base_se_norm": [-0.251870334148407, 0.0174562931060791], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.6109725832939148, -0.4613466262817383], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004541.pt"}
{"idx": 4543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S1Cn8vIGlLo.mp4", "duration": 66.83333333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the woman introduces herself as Susan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the woman introduces herself as Susan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9700748324394226, -0.8503741025924683], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.1920199990272522, -0.13216960430145264], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004543.pt"}
{"idx": 4545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9kxmWv1AV5I.mp4", "duration": 62.10370833333333, "prompt_text": "Please find the visual event described by the sentence 'When is olive oil being poured into the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is olive oil being poured into the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9677957892417908, -0.9033874273300171], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9677957892417908, -0.9033874273300171], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004545.pt"}
{"idx": 4547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9kxmWv1AV5I.mp4", "duration": 62.10370833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the penne pasta added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the penne pasta added to the pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.13048672676086426, -0.03387415409088135], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.13048672676086426, -0.03387415409088135], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004547.pt"}
{"idx": 4549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9kxmWv1AV5I.mp4", "duration": 62.10370833333333, "prompt_text": "Please find the visual event described by the sentence 'When is the finished pasta dish garnished with parsley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the finished pasta dish garnished with parsley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 53.0], "base_se_norm": [0.5135971307754517, 0.706822395324707], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.5135971307754517, 0.5780055522918701], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004549.pt"}
{"idx": 4551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CYgs0ccffVw.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the letter 'd' shown on the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the letter 'd' shown on the tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, 0.41089117527008057], "gt_se_raw": [0.0, 16.0], "gt_se_norm": [-1.0, 0.1881188154220581], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004551.pt"}
{"idx": 4553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CYgs0ccffVw.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the tablet's voice says 'E is dots one five'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the tablet's voice says 'E is dots one five'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [0.41089117527008057, 0.559406042098999], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [0.2623763084411621, 0.485148549079895], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004553.pt"}
{"idx": 4555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t3NgDZTVPR0.mp4", "duration": 133.84204166666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a woman is carrying the boy while walking on a brick path by the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a woman is carrying the boy while walking on a brick path by the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [-0.8655130863189697, -0.8356271386146545], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [-0.8804560899734497, -0.8356271386146545], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004555.pt"}
{"idx": 4557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t3NgDZTVPR0.mp4", "duration": 133.84204166666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the boy is smiling happily as his mother helps him into the toy tractor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the boy is smiling happily as his mother helps him into the toy tractor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 49.0], "base_se_norm": [-0.34250855445861816, -0.26779359579086304], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [-0.3275655508041382, -0.26779359579086304], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004557.pt"}
{"idx": 4559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBLCi5Tdpbw.mp4", "duration": 138.24, "prompt_text": "Please find the visual event described by the sentence 'When is the person folding the green paper in half for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the person folding the green paper in half for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 14.0], "base_se_norm": [-0.8987268805503845, -0.7974537014961243], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [-0.8842592835426331, -0.8119212985038757], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004559.pt"}
{"idx": 4561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBLCi5Tdpbw.mp4", "duration": 138.24, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the paper is being folded into a kite shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the paper is being folded into a kite shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 72.0], "base_se_norm": [-0.32002317905426025, 0.041666626930236816], "gt_se_raw": [46.0, 72.0], "gt_se_norm": [-0.3344907760620117, 0.041666626930236816], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004561.pt"}
{"idx": 4563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBLCi5Tdpbw.mp4", "duration": 138.24, "prompt_text": "Please find the visual event described by the sentence 'At what point is the completed paper alligator with teeth displayed outdoors by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the completed paper alligator with teeth displayed outdoors by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 124.0], "base_se_norm": [0.6348378658294678, 0.7939814329147339], "gt_se_raw": [112.0, 125.0], "gt_se_norm": [0.6203702688217163, 0.8084490299224854], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004563.pt"}
{"idx": 4565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D8Laqglo1ig.mp4", "duration": 143.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a white car parked in a disabled spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a white car parked in a disabled spot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 20.0], "base_se_norm": [-0.8192352056503296, -0.7219003438949585], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.3186558485031128, -0.27694088220596313], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004565.pt"}
{"idx": 4567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/D8Laqglo1ig.mp4", "duration": 143.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are shopping carts shown blocking an accessible path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are shopping carts shown blocking an accessible path?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.13789105415344238, -0.09617608785629272], "gt_se_raw": [61.0, 65.0], "gt_se_norm": [-0.15179604291915894, -0.09617608785629272], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004567.pt"}
{"idx": 4569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6cbRfTIrmc.mp4", "duration": 23.095325726904672, "prompt_text": "Please find the visual event described by the sentence 'When does the cat start grooming itself while lying on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat start grooming itself while lying on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 16.0], "base_se_norm": [-0.5670119524002075, 0.38556182384490967], "gt_se_raw": [4.0, 15.0], "gt_se_norm": [-0.6536095142364502, 0.29896414279937744], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004569.pt"}
{"idx": 4571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6cbRfTIrmc.mp4", "duration": 23.095325726904672, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a cat is resting on its back on a sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a cat is resting on its back on a sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, 0.6453546285629272], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.7319521903991699], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004571.pt"}
{"idx": 4573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6cbRfTIrmc.mp4", "duration": 23.095325726904672, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where a red pillow is visible next to the cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where a red pillow is visible next to the cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 19.0], "base_se_norm": [-0.6536095142364502, 0.6453546285629272], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.7319521903991699], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004573.pt"}
{"idx": 4575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PGKmexNTHNE.mp4", "duration": 12.24, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the lamp is bouncing on the letter 'I'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the lamp is bouncing on the letter 'I'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.673202633857727, -0.18300652503967285], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.3464052081108093, 0.30718958377838135], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004575.pt"}
{"idx": 4577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PGKmexNTHNE.mp4", "duration": 12.24, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the lamp settles and looks at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the lamp settles and looks at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 11.0], "base_se_norm": [0.47058820724487305, 0.7973856925964355], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [0.30718958377838135, 0.6339869499206543], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004577.pt"}
{"idx": 4579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5Ddap2Pyhtw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of the two men's faces as they begin their conversation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of the two men's faces as they begin their conversation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 21.0], "base_se_norm": [-1.0, -0.7401978969573975], "gt_se_raw": [0.0, 13.0], "gt_se_norm": [-1.0, -0.8391700983047485], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004579.pt"}
{"idx": 4581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5Ddap2Pyhtw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the blue overalls excitedly talk about tending the rabbits?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the blue overalls excitedly talk about tending the rabbits?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 82.0], "base_se_norm": [-0.6659687161445618, 0.01446545124053955], "gt_se_raw": [82.0, 85.0], "gt_se_norm": [0.01446545124053955, 0.051579952239990234], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004581.pt"}
{"idx": 4583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5Ddap2Pyhtw.mp4", "duration": 161.6615, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where the man in the jacket holds the gun to his temple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where the man in the jacket holds the gun to his temple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 123.0], "base_se_norm": [0.44746899604797363, 0.5216981172561646], "gt_se_raw": [117.0, 122.0], "gt_se_norm": [0.44746899604797363, 0.5093265771865845], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004583.pt"}
{"idx": 4585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LzKwClRvRco.mp4", "duration": 291.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the back seat look around nervously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the back seat look around nervously?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.8145179748535156, -0.7801694869995117], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.8145179748535156, -0.7801694869995117], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004585.pt"}
{"idx": 4587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LzKwClRvRco.mp4", "duration": 291.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man wearing sunglasses loads his rifle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man wearing sunglasses loads his rifle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 125.0], "base_se_norm": [-0.1893748641014099, -0.1412869095802307], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [-0.1893748641014099, -0.16876572370529175], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004587.pt"}
{"idx": 4589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TIQZjO5dfnQ.mp4", "duration": 74.0, "prompt_text": "Please find the visual event described by the sentence 'When is a person looking at themself in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person looking at themself in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.6756756901741028, -0.5675675868988037], "gt_se_raw": [14.0, 15.0], "gt_se_norm": [-0.6216216087341309, -0.5945945978164673], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004589.pt"}
{"idx": 4591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TIQZjO5dfnQ.mp4", "duration": 74.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a person is holding a framed picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a person is holding a framed picture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 42.0], "base_se_norm": [0.08108103275299072, 0.13513517379760742], "gt_se_raw": [40.0, 41.0], "gt_se_norm": [0.08108103275299072, 0.10810816287994385], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004591.pt"}
{"idx": 4593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TIQZjO5dfnQ.mp4", "duration": 74.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the 'Avengers Endgame' title card?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the 'Avengers Endgame' title card?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 63.0], "base_se_norm": [0.5675675868988037, 0.7027027606964111], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [0.5405405759811401, 0.6216216087341309], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004593.pt"}
{"idx": 4595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MmVXLM97YL0.mp4", "duration": 106.6065, "prompt_text": "Please find the visual event described by the sentence 'Show the process of mixing the batter for the strawberry cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of mixing the batter for the strawberry cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 36.0], "base_se_norm": [-0.5309854745864868, -0.3246190547943115], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.3996613621711731, -0.3433796167373657], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004595.pt"}
{"idx": 4597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MmVXLM97YL0.mp4", "duration": 106.6065, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment when frosting is being poured over the cookies and cream cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment when frosting is being poured over the cookies and cream cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [0.1819167137145996, 0.2569589614868164], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [0.2006772756576538, 0.2381983995437622], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004597.pt"}
{"idx": 4599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/avY2zI8fhUc.mp4", "duration": 144.24, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman in a pink shirt and white hat sitting in a wheelchair and speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman in a pink shirt and white hat sitting in a wheelchair and speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.9584026336669922, -0.8752080202102661], "gt_se_raw": [4.0, 10.0], "gt_se_norm": [-0.9445368647575378, -0.861342191696167], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004599.pt"}
{"idx": 4601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/avY2zI8fhUc.mp4", "duration": 144.24, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a man in sunglasses and a striped shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a man in sunglasses and a striped shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 54.0], "base_se_norm": [-0.48696619272232056, -0.25124794244766235], "gt_se_raw": [37.0, 54.0], "gt_se_norm": [-0.48696619272232056, -0.25124794244766235], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004601.pt"}
{"idx": 4603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/avY2zI8fhUc.mp4", "duration": 144.24, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when three women are at a desk communicating, with one using sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when three women are at a desk communicating, with one using sign language.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 123.0], "base_se_norm": [0.4559067487716675, 0.7054908275604248], "gt_se_raw": [106.0, 123.0], "gt_se_norm": [0.4697725772857666, 0.7054908275604248], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004603.pt"}
{"idx": 4605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZfjrd-B0ho.mp4", "duration": 163.99716666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where two people are riding a tandem bike.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where two people are riding a tandem bike.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 19.0], "base_se_norm": [-0.7926793098449707, -0.7682886719703674], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.7926793098449707, -0.7682886719703674], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004605.pt"}
{"idx": 4607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZfjrd-B0ho.mp4", "duration": 163.99716666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a person in an orange shirt is learning to fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a person in an orange shirt is learning to fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.7804840207099915, -0.7560933232307434], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.2316940426826477, -0.20730334520339966], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004607.pt"}
{"idx": 4609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hWskc5_mPSI.mp4", "duration": 14.948384501943961, "prompt_text": "Please find the visual event described by the sentence 'When are two girls seen laughing together on the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two girls seen laughing together on the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.732412576675415, -0.4648250937461853], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.732412576675415, -0.5986188650131226], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004609.pt"}
{"idx": 4611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hWskc5_mPSI.mp4", "duration": 14.948384501943961, "prompt_text": "Please find the visual event described by the sentence 'At what point is a green flag with a yellow 'A' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a green flag with a yellow 'A' visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [0.33793723583221436, 0.8731120824813843], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [0.33793723583221436, 0.7393183708190918], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004611.pt"}
{"idx": 4613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hWskc5_mPSI.mp4", "duration": 14.948384501943961, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where children's voices are heard shouting 'A'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where children's voices are heard shouting 'A'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.732412576675415, -0.4648250937461853], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8662062883377075], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004613.pt"}
{"idx": 4615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bCO3S2_s07w.mp4", "duration": 176.20936666666668, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the woman place her hand on the boy's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the woman place her hand on the boy's shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 136.0], "base_se_norm": [0.3847163915634155, 0.5436183214187622], "gt_se_raw": [119.0, 121.0], "gt_se_norm": [0.3506660461425781, 0.37336623668670654], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004615.pt"}
{"idx": 4617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bCO3S2_s07w.mp4", "duration": 176.20936666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the woman take the device and look at it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman take the device and look at it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 147.0], "base_se_norm": [0.5663185119628906, 0.6684697866439819], "gt_se_raw": [141.0, 153.0], "gt_se_norm": [0.6003689765930176, 0.7365705966949463], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004617.pt"}
{"idx": 4619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CGAB2warrpA.mp4", "duration": 120.98753333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the red shirt trip over a rock and fall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the red shirt trip over a rock and fall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.5371423959732056, -0.4710198640823364], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.5206117630004883, -0.4710198640823364], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004619.pt"}
{"idx": 4621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CGAB2warrpA.mp4", "duration": 120.98753333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman open the door for a child in a purple outfit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman open the door for a child in a purple outfit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.04142963886260986, 0.12408280372619629], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [0.05796027183532715, 0.107552170753479], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004621.pt"}
{"idx": 4623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CGAB2warrpA.mp4", "duration": 120.98753333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a girl in a purple dress helps her mother carry boxes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a girl in a purple dress helps her mother carry boxes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 98.0], "base_se_norm": [0.45469534397125244, 0.6200016736984253], "gt_se_raw": [93.0, 98.0], "gt_se_norm": [0.5373485088348389, 0.6200016736984253], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004623.pt"}
{"idx": 4625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iFIsvjt9NSs.mp4", "duration": 176.00916666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the kid is playing with the toy car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the kid is playing with the toy car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 162.0], "base_se_norm": [-0.1818608045578003, 0.840813159942627], "gt_se_raw": [73.0, 76.0], "gt_se_norm": [-0.17049777507781982, -0.13640862703323364], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004625.pt"}
{"idx": 4627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iFIsvjt9NSs.mp4", "duration": 176.00916666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the infant throw a colorful toy on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the infant throw a colorful toy on the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.8295543193817139, -0.7954652309417725], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.8181912899017334, -0.7841021418571472], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004627.pt"}
{"idx": 4629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uqxzi_ghjgc.mp4", "duration": 110.0, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman dancing while cooking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman dancing while cooking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 23.0], "base_se_norm": [-0.8727272748947144, -0.5818182229995728], "gt_se_raw": [8.0, 16.0], "gt_se_norm": [-0.8545454740524292, -0.7090909481048584], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004629.pt"}
{"idx": 4631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uqxzi_ghjgc.mp4", "duration": 110.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the mother hand the violin to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the mother hand the violin to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 51.0], "base_se_norm": [-0.12727272510528564, -0.0727272629737854], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.12727272510528564, -0.0727272629737854], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004631.pt"}
{"idx": 4633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uqxzi_ghjgc.mp4", "duration": 110.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the mother discovers and opens a secret passage behind the bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the mother discovers and opens a secret passage behind the bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 78.0], "base_se_norm": [0.20000004768371582, 0.41818177700042725], "gt_se_raw": [70.0, 76.0], "gt_se_norm": [0.27272725105285645, 0.38181817531585693], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004633.pt"}
{"idx": 4635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uYOpUXb8tYU.mp4", "duration": 184.4843, "prompt_text": "Please find the visual event described by the sentence 'When does the woman gesture towards the model of the space shuttle stack while explaining its components?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman gesture towards the model of the space shuttle stack while explaining its components?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 68.0], "base_se_norm": [-0.3712202310562134, -0.2628098726272583], "gt_se_raw": [58.0, 70.0], "gt_se_norm": [-0.3712202310562134, -0.2411278486251831], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004635.pt"}
{"idx": 4637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uYOpUXb8tYU.mp4", "duration": 184.4843, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when an animated space shuttle flies into the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when an animated space shuttle flies into the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.8699076175689697, -0.8265435099601746], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.8590666055679321, -0.8265435099601746], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004637.pt"}
{"idx": 4639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-l0tT9PVadQ.mp4", "duration": 126.29283333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the person fold the sides of the paper towards the center crease?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person fold the sides of the paper towards the center crease?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 38.0], "base_se_norm": [-0.6832757592201233, -0.3982239365577698], "gt_se_raw": [20.0, 31.0], "gt_se_norm": [-0.6832757592201233, -0.5090774297714233], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004639.pt"}
{"idx": 4641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-l0tT9PVadQ.mp4", "duration": 126.29283333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the person apply glue to the paper before rolling it into a cone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person apply glue to the paper before rolling it into a cone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 64.0], "base_se_norm": [-0.04982727766036987, 0.013517498970031738], "gt_se_raw": [59.0, 64.0], "gt_se_norm": [-0.06566351652145386, 0.013517498970031738], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004641.pt"}
{"idx": 4643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-l0tT9PVadQ.mp4", "duration": 126.29283333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the person attaches the eyes to the fish's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the person attaches the eyes to the fish's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [0.7261470556259155, 0.7894918918609619], "gt_se_raw": [110.0, 113.0], "gt_se_norm": [0.7419832944869995, 0.7894918918609619], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004643.pt"}
{"idx": 4645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4h-PtwiqWE.mp4", "duration": 9.8, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time the girl presses down on the middle of the board with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time the girl presses down on the middle of the board with both hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.3877550959587097, 0.22448980808258057], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.18367350101470947, 0.020408153533935547], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004645.pt"}
{"idx": 4647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4h-PtwiqWE.mp4", "duration": 9.8, "prompt_text": "Please find the visual event described by the sentence 'Show me when the girl is touching the bottom black strip on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the girl is touching the bottom black strip on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.18367350101470947, 0.42857134342193604], "gt_se_raw": [5.0, 7.0], "gt_se_norm": [0.020408153533935547, 0.42857134342193604], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004647.pt"}
{"idx": 4649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OINa46HeWg8.mp4", "duration": 130.04657777777777, "prompt_text": "Please find the visual event described by the sentence 'When can we see a couple cuddling in bed in the morning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a couple cuddling in bed in the morning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.9692417979240417, -0.8308298587799072], "gt_se_raw": [1.0, 11.0], "gt_se_norm": [-0.9846208691596985, -0.8308298587799072], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004649.pt"}
{"idx": 4651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OINa46HeWg8.mp4", "duration": 130.04657777777777, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man proposes to a woman on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man proposes to a woman on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 49.0], "base_se_norm": [-0.33869850635528564, -0.2464239001274109], "gt_se_raw": [44.0, 49.0], "gt_se_norm": [-0.3233194351196289, -0.2464239001274109], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004651.pt"}
{"idx": 4653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OINa46HeWg8.mp4", "duration": 130.04657777777777, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a group of people are singing 'Happy Birthday' around a cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a group of people are singing 'Happy Birthday' around a cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 109.0], "base_se_norm": [0.3841193914413452, 0.676322340965271], "gt_se_raw": [91.0, 109.0], "gt_se_norm": [0.39949846267700195, 0.676322340965271], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004653.pt"}
{"idx": 4655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTMKUe8VF0M.mp4", "duration": 401.4343666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the woman in the black shirt guides the boy's hand towards the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the woman in the black shirt guides the boy's hand towards the toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 27.0], "base_se_norm": [-0.9352322816848755, -0.8654823303222656], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.9103215932846069, -0.9003573060035706], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004655.pt"}
{"idx": 4657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UTMKUe8VF0M.mp4", "duration": 401.4343666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the camouflage shirt give the boy a high-five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 288.0 - 294.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the camouflage shirt give the boy a high-five?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 288.0 - 294.0 seconds.<|im_end|>\n", "base_se_raw": [288.0, 294.0], "base_se_norm": [0.43485474586486816, 0.4647475481033325], "gt_se_raw": [288.0, 290.0], "gt_se_norm": [0.43485474586486816, 0.4448190927505493], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004657.pt"}
{"idx": 4659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zE0qp3Yje6Q.mp4", "duration": 170.16, "prompt_text": "Please find the visual event described by the sentence 'When can we see a turtle swimming gracefully in the deep blue ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a turtle swimming gracefully in the deep blue ocean?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 31.0], "base_se_norm": [-0.976492702960968, -0.6356370449066162], "gt_se_raw": [2.0, 12.0], "gt_se_norm": [-0.976492702960968, -0.8589562773704529], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004659.pt"}
{"idx": 4661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zE0qp3Yje6Q.mp4", "duration": 170.16, "prompt_text": "Please find the visual event described by the sentence 'At what point does a baby turtle first emerge from the sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a baby turtle first emerge from the sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 56.0], "base_se_norm": [-0.4475787878036499, -0.34179598093032837], "gt_se_raw": [48.0, 54.0], "gt_se_norm": [-0.4358251094818115, -0.36530327796936035], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004661.pt"}
{"idx": 4663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zE0qp3Yje6Q.mp4", "duration": 170.16, "prompt_text": "Please find the visual event described by the sentence 'Show me the baby turtles struggling against the waves as they enter the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the baby turtles struggling against the waves as they enter the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 108.0], "base_se_norm": [0.0460742712020874, 0.2693934440612793], "gt_se_raw": [118.0, 122.0], "gt_se_norm": [0.38692986965179443, 0.4339444637298584], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004663.pt"}
{"idx": 4665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgfTgj_muvw.mp4", "duration": 130.4637137137137, "prompt_text": "Please find the visual event described by the sentence 'When are the man and woman in wheelchairs introduced together on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and woman in wheelchairs introduced together on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 33.0], "base_se_norm": [-0.5707619190216064, -0.49411219358444214], "gt_se_raw": [27.0, 33.0], "gt_se_norm": [-0.5860918164253235, -0.49411219358444214], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004665.pt"}
{"idx": 4667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zgfTgj_muvw.mp4", "duration": 130.4637137137137, "prompt_text": "Please find the visual event described by the sentence 'When are the apps specifically for visually impaired users, like LookTel and TapTapSee, mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the apps specifically for visually impaired users, like LookTel and TapTapSee, mentioned?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 73.0], "base_se_norm": [0.011775612831115723, 0.11908507347106934], "gt_se_raw": [67.0, 73.0], "gt_se_norm": [0.02710545063018799, 0.11908507347106934], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004667.pt"}
{"idx": 4669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ND63UPelkIw.mp4", "duration": 143.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog use a rope to open the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog use a rope to open the refrigerator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 64.0], "base_se_norm": [-0.19407135248184204, -0.11069941520690918], "gt_se_raw": [58.0, 65.0], "gt_se_norm": [-0.19407135248184204, -0.09680408239364624], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004669.pt"}
{"idx": 4671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ND63UPelkIw.mp4", "duration": 143.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the red shirt walk with a walker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the red shirt walk with a walker?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 26.0], "base_se_norm": [-0.7915701866149902, -0.6387216448783875], "gt_se_raw": [15.0, 25.0], "gt_se_norm": [-0.7915701866149902, -0.6526169776916504], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004671.pt"}
{"idx": 4673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ND63UPelkIw.mp4", "duration": 143.93333333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the clip where the man falls out of his wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the clip where the man falls out of his wheelchair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 39.0], "base_se_norm": [-0.4997684359550476, -0.4580824375152588], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [-0.4997684359550476, -0.4580824375152588], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004673.pt"}
{"idx": 4675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EIYDPwY3gNk.mp4", "duration": 85.96, "prompt_text": "Please find the visual event described by the sentence 'A man in a red sweater enters the room and startles the woman. Please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a red sweater enters the room and startles the woman. Please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 14.0], "base_se_norm": [-0.8604001998901367, -0.674267053604126], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.8604001998901367, -0.7906002998352051], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004675.pt"}
{"idx": 4677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EIYDPwY3gNk.mp4", "duration": 85.96, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man puts his arm around the woman to watch the movie together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man puts his arm around the woman to watch the movie together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [0.09353184700012207, 0.20986509323120117], "gt_se_raw": [47.0, 51.0], "gt_se_norm": [0.09353184700012207, 0.18659842014312744], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004677.pt"}
{"idx": 4679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gIUSfQ34Hj4.mp4", "duration": 161.3612, "prompt_text": "Please find the visual event described by the sentence 'When is shampoo poured into the glass bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is shampoo poured into the glass bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8636599183082581, -0.8016871213912964], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.8636599183082581, -0.8140817284584045], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004679.pt"}
{"idx": 4681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gIUSfQ34Hj4.mp4", "duration": 161.3612, "prompt_text": "Please find the visual event described by the sentence 'At what moment is the mixture stirred with a red spoon after the salt is added?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is the mixture stirred with a red spoon after the salt is added?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 73.0], "base_se_norm": [-0.28111594915390015, -0.09519761800765991], "gt_se_raw": [58.0, 74.0], "gt_se_norm": [-0.28111594915390015, -0.08280307054519653], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004681.pt"}
{"idx": 4683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gIUSfQ34Hj4.mp4", "duration": 161.3612, "prompt_text": "Please find the visual event described by the sentence 'When can we see the slime being stretched out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the slime being stretched out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 126.0], "base_se_norm": [0.4501626491546631, 0.561713695526123], "gt_se_raw": [118.0, 127.0], "gt_se_norm": [0.46255719661712646, 0.5741082429885864], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004683.pt"}
{"idx": 4685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFJAuAWxuvI.mp4", "duration": 178.57840000000002, "prompt_text": "Please find the visual event described by the sentence 'At what point are the egg and vanilla extract being beaten together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the egg and vanilla extract being beaten together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 56.0], "base_se_norm": [-0.4400218725204468, -0.372824490070343], "gt_se_raw": [50.0, 56.0], "gt_se_norm": [-0.4400218725204468, -0.372824490070343], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004685.pt"}
{"idx": 4687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PFJAuAWxuvI.mp4", "duration": 178.57840000000002, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for shaping the cookie dough into balls?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for shaping the cookie dough into balls?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 128.0], "base_se_norm": [0.2655506134033203, 0.4335440397262573], "gt_se_raw": [113.0, 121.0], "gt_se_norm": [0.2655506134033203, 0.355147123336792], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004687.pt"}
{"idx": 4689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5lgHJB1lwYs.mp4", "duration": 68.36828495161828, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman speaking in a red shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman speaking in a red shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 17.0], "base_se_norm": [-0.9414933323860168, -0.5026934146881104], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.9414933323860168, -0.7952266931533813], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004689.pt"}
{"idx": 4691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5lgHJB1lwYs.mp4", "duration": 68.36828495161828, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a woman is doing a yoga pose on a porch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a woman is doing a yoga pose on a porch.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [-0.23941344022750854, -0.06389343738555908], "gt_se_raw": [26.0, 32.0], "gt_se_norm": [-0.23941344022750854, -0.06389343738555908], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004691.pt"}
{"idx": 4693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5lgHJB1lwYs.mp4", "duration": 68.36828495161828, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the UWF Recreation Facebook page with upcoming events listed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the UWF Recreation Facebook page with upcoming events listed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 62.0], "base_se_norm": [0.6674398183822632, 0.8137063980102539], "gt_se_raw": [57.0, 61.0], "gt_se_norm": [0.6674398183822632, 0.7844531536102295], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004693.pt"}
{"idx": 4695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UhQ9HquDNEM.mp4", "duration": 137.08, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man presents the catnip to the cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man presents the catnip to the cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.7519696354866028, -0.6644295454025269], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.7519696354866028, -0.6644295454025269], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004695.pt"}
{"idx": 4697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UhQ9HquDNEM.mp4", "duration": 137.08, "prompt_text": "Please find the visual event described by the sentence 'At what point does the cat tear the pouch, causing the catnip to spill everywhere?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the cat tear the pouch, causing the catnip to spill everywhere?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 100.0], "base_se_norm": [0.31310176849365234, 0.4590020179748535], "gt_se_raw": [90.0, 92.0], "gt_se_norm": [0.31310176849365234, 0.3422818183898926], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004697.pt"}
{"idx": 4699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-u_KZ1BCaTw.mp4", "duration": 160.52703333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see the mobile app interface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the mobile app interface?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [-0.8255745768547058, -0.7757387161254883], "gt_se_raw": [14.0, 18.0], "gt_se_norm": [-0.8255745768547058, -0.7757387161254883], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004699.pt"}
{"idx": 4701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-u_KZ1BCaTw.mp4", "duration": 160.52703333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point is the exterior of the Redfern station shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the exterior of the Redfern station shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 53.0], "base_se_norm": [-0.40196990966796875, -0.3396751284599304], "gt_se_raw": [48.0, 53.0], "gt_se_norm": [-0.40196990966796875, -0.3396751284599304], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004701.pt"}
{"idx": 4703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-u_KZ1BCaTw.mp4", "duration": 160.52703333333335, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the text 'SECTION 24' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the text 'SECTION 24' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 139.0], "base_se_norm": [0.6819596290588379, 0.7317954301834106], "gt_se_raw": [134.0, 139.0], "gt_se_norm": [0.6695007085800171, 0.7317954301834106], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004703.pt"}
{"idx": 4705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebyf6FzKEiI.mp4", "duration": 115.2151, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the baby in the purple outfit lifts its head and looks up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the baby in the purple outfit lifts its head and looks up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 48.0], "base_se_norm": [-0.3924407362937927, -0.1667758822441101], "gt_se_raw": [36.0, 40.0], "gt_se_norm": [-0.37508195638656616, -0.3056465983390808], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004705.pt"}
{"idx": 4707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ebyf6FzKEiI.mp4", "duration": 115.2151, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman is holding and playing with the child's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman is holding and playing with the child's fingers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 70.0], "base_se_norm": [-0.4792349338531494, 0.21511852741241455], "gt_se_raw": [91.0, 98.0], "gt_se_norm": [0.5796539783477783, 0.7011659145355225], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004707.pt"}
{"idx": 4709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nG2rNBFzkGE.mp4", "duration": 16.616600000000002, "prompt_text": "Please find the visual event described by the sentence 'When does the cat yawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat yawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.7592768669128418], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8796384334564209], "start_token_span": [61, 64], "end_token_span": [66, 69], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004709.pt"}
{"idx": 4711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nG2rNBFzkGE.mp4", "duration": 16.616600000000002, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the cat leaps off the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the cat leaps off the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [0.3239772319793701, 0.5647003650665283], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.3239772319793701, 0.5647003650665283], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004711.pt"}
{"idx": 4713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nG2rNBFzkGE.mp4", "duration": 16.616600000000002, "prompt_text": "Please find the visual event described by the sentence 'When is the orange and white cat visible in the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the orange and white cat visible in the frame?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, 0.5647003650665283], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, 0.4443387985229492], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004713.pt"}
{"idx": 4715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uODXWDOj0ow.mp4", "duration": 30.797433333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man in a striped shirt is wiring a control panel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man in a striped shirt is wiring a control panel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4804762005805969, -0.09083336591720581], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5454167127609253, -0.09083336591720581], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004715.pt"}
{"idx": 4717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uODXWDOj0ow.mp4", "duration": 30.797433333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up of a woman smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up of a woman smiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.42869043350219727, 0.8183332681655884], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.42869043350219727, 0.8183332681655884], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004717.pt"}
{"idx": 4719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/osP9iJjvlAE.mp4", "duration": 159.20070833333332, "prompt_text": "Please find the visual event described by the sentence 'When does the woman leap off the bridge into the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman leap off the bridge into the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 32.0], "base_se_norm": [-0.6231172680854797, -0.5979917645454407], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.6231172680854797, -0.5979917645454407], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004719.pt"}
{"idx": 4721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/osP9iJjvlAE.mp4", "duration": 159.20070833333332, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a full moon in the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a full moon in the night sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 17.0], "base_se_norm": [-0.8115586042404175, -0.7864331007003784], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.8115586042404175, -0.7864331007003784], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004721.pt"}
{"idx": 4723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/osP9iJjvlAE.mp4", "duration": 159.20070833333332, "prompt_text": "Please find the visual event described by the sentence 'When does the woman defiantly state, 'Don't you tell me what I can't do'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman defiantly state, 'Don't you tell me what I can't do'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 31.0], "base_se_norm": [-0.6482427716255188, -0.6105545163154602], "gt_se_raw": [94.0, 96.0], "gt_se_norm": [0.1808992624282837, 0.20602476596832275], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004723.pt"}
{"idx": 4725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x-wcNG_Njk8.mp4", "duration": 126.29283333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point are the T-shirts being advertised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the T-shirts being advertised?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 106.0], "base_se_norm": [0.47276771068573, 0.6786384582519531], "gt_se_raw": [94.0, 106.0], "gt_se_norm": [0.4886038303375244, 0.6786384582519531], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004725.pt"}
{"idx": 4727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/x-wcNG_Njk8.mp4", "duration": 126.29283333333332, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a blurry red circle on a black background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a blurry red circle on a black background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.8258016705513, -0.7941292524337769], "gt_se_raw": [40.0, 41.0], "gt_se_norm": [-0.3665515184402466, -0.3507153391838074], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004727.pt"}
{"idx": 4729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RfI7lW6MbEU.mp4", "duration": 21.788433333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time when a man in a blue T-shirt is giving a lecture to a group of people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when a man in a blue T-shirt is giving a lecture to a group of people.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, -0.5410408973693848], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.5410408973693848], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004729.pt"}
{"idx": 4731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RfI7lW6MbEU.mp4", "duration": 21.788433333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman in a striped apron is stirring something in a bowl on a counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman in a striped apron is stirring something in a bowl on a counter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.08208179473876953, 0.19329369068145752], "gt_se_raw": [10.0, 13.0], "gt_se_norm": [-0.08208179473876953, 0.19329369068145752], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004731.pt"}
{"idx": 4733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RfI7lW6MbEU.mp4", "duration": 21.788433333333334, "prompt_text": "Please find the visual event described by the sentence 'When is a group of people seen gathered in a kitchen, with some cooking and others talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a group of people seen gathered in a kitchen, with some cooking and others talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [0.5604610443115234, 0.8358364105224609], "gt_se_raw": [17.0, 21.0], "gt_se_norm": [0.5604610443115234, 0.9276282787322998], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004733.pt"}
{"idx": 4735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZPkuz0gRPU.mp4", "duration": 156.86504166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman is working in an additive manufacturing lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman is working in an additive manufacturing lab.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 65.0], "base_se_norm": [-0.19676172733306885, -0.17126208543777466], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.19676172733306885, -0.17126208543777466], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004735.pt"}
{"idx": 4737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZPkuz0gRPU.mp4", "duration": 156.86504166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a blonde woman is browsing a website on her phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a blonde woman is browsing a website on her phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 122.0], "base_se_norm": [0.44072890281677246, 0.5554772615432739], "gt_se_raw": [113.0, 122.0], "gt_se_norm": [0.44072890281677246, 0.5554772615432739], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004737.pt"}
{"idx": 4739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFinNxS5KN4.mp4", "duration": 160.11829166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the 'Jurassic World' gate opening?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the 'Jurassic World' gate opening?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.5628231763839722, -0.5253509283065796], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.5628231763839722, -0.5253509283065796], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004739.pt"}
{"idx": 4741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFinNxS5KN4.mp4", "duration": 160.11829166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a man rides a motorcycle alongside dinosaurs in the dark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a man rides a motorcycle alongside dinosaurs in the dark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 145.0], "base_se_norm": [0.7861795425415039, 0.8111610412597656], "gt_se_raw": [142.0, 145.0], "gt_se_norm": [0.7736886739730835, 0.8111610412597656], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004741.pt"}
{"idx": 4743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RFinNxS5KN4.mp4", "duration": 160.11829166666666, "prompt_text": "Please find the visual event described by the sentence 'When are the two boys riding in the gyrosphere next to a large dinosaur?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two boys riding in the gyrosphere next to a large dinosaur?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.3629709482192993, -0.30051714181900024], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [-0.35048019886016846, -0.30051714181900024], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004743.pt"}
{"idx": 4745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fDDw7W2RQYo.mp4", "duration": 150.98416666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is flying a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is flying a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 81.0], "base_se_norm": [-0.006518304347991943, 0.0729602575302124], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [0.006728172302246094, 0.0729602575302124], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004745.pt"}
{"idx": 4747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fDDw7W2RQYo.mp4", "duration": 150.98416666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man is buried in the sand with only his head visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man is buried in the sand with only his head visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 116.0], "base_se_norm": [0.43061363697052, 0.5365849733352661], "gt_se_raw": [107.0, 113.0], "gt_se_norm": [0.41736721992492676, 0.49684572219848633], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004747.pt"}
{"idx": 4749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CzNb3zMq6C4.mp4", "duration": 185.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'When do the members hold up signs with 'J-HOPE' on them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the members hold up signs with 'J-HOPE' on them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.9030520915985107, -0.8707361221313477], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.9030520915985107, -0.8707361221313477], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004749.pt"}
{"idx": 4751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CzNb3zMq6C4.mp4", "duration": 185.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the majority of members hold up a sign that says 'JUNGKOOK'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the majority of members hold up a sign that says 'JUNGKOOK'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 89.0], "base_se_norm": [-0.09515261650085449, -0.041292667388916016], "gt_se_raw": [85.0, 89.0], "gt_se_norm": [-0.0843806266784668, -0.041292667388916016], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004751.pt"}
{"idx": 4753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CzNb3zMq6C4.mp4", "duration": 185.66666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the timeframe where the majority of members are showing signs that say 'JIMIN'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the timeframe where the majority of members are showing signs that say 'JIMIN'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 57.0], "base_se_norm": [-0.5044883489608765, -0.38599640130996704], "gt_se_raw": [128.0, 131.0], "gt_se_norm": [0.37881505489349365, 0.41113102436065674], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004753.pt"}
{"idx": 4755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsOzAbUt8n8.mp4", "duration": 124.68, "prompt_text": "Please find the visual event described by the sentence 'Show me the green and yellow jeep with people inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the green and yellow jeep with people inside.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 39.0], "base_se_norm": [-0.5027269721031189, -0.37439846992492676], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.5027269721031189, -0.4385627508163452], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004755.pt"}
{"idx": 4757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MsOzAbUt8n8.mp4", "duration": 124.68, "prompt_text": "Please find the visual event described by the sentence 'When do two people start zip-lining side by side over the green valley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two people start zip-lining side by side over the green valley?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [0.010587096214294434, 0.07475137710571289], "gt_se_raw": [63.0, 67.0], "gt_se_norm": [0.010587096214294434, 0.07475137710571289], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004757.pt"}
{"idx": 4759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9fjEsxhmQi0.mp4", "duration": 79.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman wearing a black dress and high heels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman wearing a black dress and high heels?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 79.0], "base_se_norm": [-0.9749895930290222, 0.9758232831954956], "gt_se_raw": [1.0, 79.0], "gt_se_norm": [-0.9749895930290222, 0.9758232831954956], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004759.pt"}
{"idx": 4761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9fjEsxhmQi0.mp4", "duration": 79.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does a sun-like halo appear behind the woman's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a sun-like halo appear behind the woman's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 79.0], "base_se_norm": [-0.9749895930290222, 0.9758232831954956], "gt_se_raw": [1.0, 79.0], "gt_se_norm": [-0.9749895930290222, 0.9758232831954956], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004761.pt"}
{"idx": 4763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9fjEsxhmQi0.mp4", "duration": 79.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the abstract blurry shape is shown at the very beginning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 1.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the abstract blurry shape is shown at the very beginning.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 1.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 1.0], "base_se_norm": [-1.0, -0.9749895930290222], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.9749895930290222], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004763.pt"}
{"idx": 4765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5bWPhOp8TCw.mp4", "duration": 186.25273333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman gets upset, talking about how her friends matured overnight and she didn't.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman gets upset, talking about how her friends matured overnight and she didn't.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 100.0], "base_se_norm": [-0.48457133769989014, 0.07380974292755127], "gt_se_raw": [47.0, 65.0], "gt_se_norm": [-0.49530941247940063, -0.30202364921569824], "start_token_span": [76, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004765.pt"}
{"idx": 4767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5bWPhOp8TCw.mp4", "duration": 186.25273333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman gets angry and asks the man if he takes a drug to be so happy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman gets angry and asks the man if he takes a drug to be so happy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [-0.3664522171020508, -0.30202364921569824], "gt_se_raw": [85.0, 91.0], "gt_se_norm": [-0.08726173639297485, -0.02283310890197754], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004767.pt"}
{"idx": 4769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d-84UJpYFRM.mp4", "duration": 198.16463333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the hearing range of different animals shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the hearing range of different animals shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 34.0], "base_se_norm": [-0.7274993062019348, -0.6568509340286255], "gt_se_raw": [27.0, 34.0], "gt_se_norm": [-0.7274993062019348, -0.6568509340286255], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004769.pt"}
{"idx": 4771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d-84UJpYFRM.mp4", "duration": 198.16463333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the rat is shown inside a red tube.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the rat is shown inside a red tube.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 19.0], "base_se_norm": [-0.8687959909439087, -0.8082402348518372], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.8687959909439087, -0.8385180830955505], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004771.pt"}
{"idx": 4773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d-84UJpYFRM.mp4", "duration": 198.16463333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the somatosensory cortex of the rat's brain being displayed and labeled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the somatosensory cortex of the rat's brain being displayed and labeled?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 118.0], "base_se_norm": [0.009261846542358398, 0.19092893600463867], "gt_se_raw": [102.0, 109.0], "gt_se_norm": [0.029447078704833984, 0.10009539127349854], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004773.pt"}
{"idx": 4775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gj6V-xZgtlQ.mp4", "duration": 319.7083333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the grandpa and boy rafting down the river with crocodiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the grandpa and boy rafting down the river with crocodiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 162.0], "base_se_norm": [-0.036621928215026855, 0.013423681259155273], "gt_se_raw": [155.0, 165.0], "gt_se_norm": [-0.030366241931915283, 0.032190799713134766], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004775.pt"}
{"idx": 4777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Gj6V-xZgtlQ.mp4", "duration": 319.7083333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment where the boy and his teddy bear jump out of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 198.0 - 204.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment where the boy and his teddy bear jump out of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 198.0 - 204.0 seconds.<|im_end|>\n", "base_se_raw": [198.0, 204.0], "base_se_norm": [0.23862886428833008, 0.27616310119628906], "gt_se_raw": [199.0, 202.0], "gt_se_norm": [0.24488461017608643, 0.2636517286300659], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004777.pt"}
{"idx": 4779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q-FHWX1es0o.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker state that the employment rate for Americans without disabilities is higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker state that the employment rate for Americans without disabilities is higher?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9030095934867859, -0.6120384335517883], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004779.pt"}
{"idx": 4781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q-FHWX1es0o.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker mention that people with disabilities are employed at nearly half the rate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker mention that people with disabilities are employed at nearly half the rate?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.12708652019500732, 0.16388463973999023], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004781.pt"}
{"idx": 4783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q-FHWX1es0o.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'When is the data for the year 2010 shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the data for the year 2010 shown on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [77, 80], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004783.pt"}
{"idx": 4785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9lZlt-SlABw.mp4", "duration": 100.3002, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman demonstrates the vibrating bumblebee toy on the vest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman demonstrates the vibrating bumblebee toy on the vest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [-0.2622148394584656, -0.08275359869003296], "gt_se_raw": [37.0, 47.0], "gt_se_norm": [-0.2622148394584656, -0.06281346082687378], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004785.pt"}
{"idx": 4787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9lZlt-SlABw.mp4", "duration": 100.3002, "prompt_text": "Please find the visual event described by the sentence 'Show me when the speaker talks about a child being actively engaged with something when their mom is busy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the speaker talks about a child being actively engaged with something when their mom is busy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.6011972427368164, -0.40179580450057983], "gt_se_raw": [26.0, 31.0], "gt_se_norm": [-0.48155635595321655, -0.38185566663742065], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004787.pt"}
{"idx": 4789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TdUsyXQ8Wrs.mp4", "duration": 115.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a crowd of people running out of the movie theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a crowd of people running out of the movie theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 46.0], "base_se_norm": [-0.39411425590515137, -0.20369303226470947], "gt_se_raw": [34.0, 41.0], "gt_se_norm": [-0.41142529249191284, -0.2902480959892273], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004789.pt"}
{"idx": 4791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TdUsyXQ8Wrs.mp4", "duration": 115.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a meteor streaks across the night sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a meteor streaks across the night sky.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 61.0], "base_se_norm": [0.02135026454925537, 0.05597233772277832], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.02135026454925537, 0.07328331470489502], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004791.pt"}
{"idx": 4793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TdUsyXQ8Wrs.mp4", "duration": 115.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the white shirt explain what he saw to the police officer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the white shirt explain what he saw to the police officer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.7576457262039185, -0.6710906028747559], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.740334689617157, -0.6537795662879944], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004793.pt"}
{"idx": 4795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PquDhG2AhwM.mp4", "duration": 47.347300000000004, "prompt_text": "Please find the visual event described by the sentence 'Show the moment an adult's hand with a watch becomes visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment an adult's hand with a watch becomes visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.2819020748138428, -0.1129378080368042], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.1974199414253235, -0.07069677114486694], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004795.pt"}
{"idx": 4797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PquDhG2AhwM.mp4", "duration": 47.347300000000004, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the adult shakes the pom-pom while it is on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the adult shakes the pom-pom while it is on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 46.0], "base_se_norm": [0.1405085325241089, 0.9430885314941406], "gt_se_raw": [31.0, 36.0], "gt_se_norm": [0.3094726800918579, 0.5206780433654785], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004797.pt"}
{"idx": 4799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUjV78Kj9kQ.mp4", "duration": 87.52076666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Snoopy set up the table and chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Snoopy set up the table and chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 16.0], "base_se_norm": [-0.9542965888977051, -0.6343724727630615], "gt_se_raw": [6.0, 16.0], "gt_se_norm": [-0.8628896474838257, -0.6343724727630615], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004799.pt"}
{"idx": 4801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUjV78Kj9kQ.mp4", "duration": 87.52076666666667, "prompt_text": "Please find the visual event described by the sentence 'At what time are the dog and the bird eating together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the dog and the bird eating together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 70.0], "base_se_norm": [0.18828940391540527, 0.5996203422546387], "gt_se_raw": [52.0, 67.0], "gt_se_norm": [0.18828940391540527, 0.5310652256011963], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004801.pt"}
{"idx": 4803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUjV78Kj9kQ.mp4", "duration": 87.52076666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the roasted turkey first revealed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the roasted turkey first revealed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.4744104743003845, -0.40585529804229736], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.4515587091445923, -0.40585529804229736], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004803.pt"}
{"idx": 4805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-bCmuE3KxWI.mp4", "duration": 108.5, "prompt_text": "Please find the visual event described by the sentence 'Show me the process of adding flour to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the process of adding flour to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 49.0], "base_se_norm": [-0.5760368704795837, -0.09677422046661377], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.5760368704795837, -0.4838709831237793], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004805.pt"}
{"idx": 4807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-bCmuE3KxWI.mp4", "duration": 108.5, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when chocolate chips are added to the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when chocolate chips are added to the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 65.0], "base_se_norm": [0.10599076747894287, 0.1981567144393921], "gt_se_raw": [60.0, 65.0], "gt_se_norm": [0.10599076747894287, 0.1981567144393921], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004807.pt"}
{"idx": 4809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dTqW89AjVA0.mp4", "duration": 68.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy high-five the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy high-five the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9416342377662659], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.9708171486854553], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004809.pt"}
{"idx": 4811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dTqW89AjVA0.mp4", "duration": 68.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy start writing 'T' on the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy start writing 'T' on the paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [0.07976651191711426, 0.22568094730377197], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.09533071517944336, 0.021400809288024902], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004811.pt"}
{"idx": 4813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dTqW89AjVA0.mp4", "duration": 68.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher take the finished worksheet from the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher take the finished worksheet from the desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [0.8093385696411133, 0.8968871831893921], "gt_se_raw": [60.0, 62.0], "gt_se_norm": [0.7509727478027344, 0.8093385696411133], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004813.pt"}
{"idx": 4815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MJesmjQXqno.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where 'CAREERS' is written on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where 'CAREERS' is written on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.5737595558166504, -0.2540792226791382], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.5737595558166504, -0.2540792226791382], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004815.pt"}
{"idx": 4817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MJesmjQXqno.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'When does the surgeon pass a surgical instrument to the other person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the surgeon pass a surgical instrument to the other person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.5737595558166504, -0.3606393337249756], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [0.17216122150421143, 0.43856143951416016], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004817.pt"}
{"idx": 4819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5WeiYHnvNY.mp4", "duration": 161.7, "prompt_text": "Please find the visual event described by the sentence 'When are the two women cooking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two women cooking in the kitchen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.146567702293396, -0.09709334373474121], "gt_se_raw": [69.0, 73.0], "gt_se_norm": [-0.146567702293396, -0.09709334373474121], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004819.pt"}
{"idx": 4821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5WeiYHnvNY.mp4", "duration": 161.7, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a group of kids dancing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a group of kids dancing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 77.0], "base_se_norm": [-0.09709334373474121, -0.0476190447807312], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [-0.09709334373474121, -0.03525042533874512], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004821.pt"}
{"idx": 4823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5WeiYHnvNY.mp4", "duration": 161.7, "prompt_text": "Please find the visual event described by the sentence 'Locate the woman wearing a red and black plaid jacket while she is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the woman wearing a red and black plaid jacket while she is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 108.0], "base_se_norm": [0.21212124824523926, 0.33580708503723145], "gt_se_raw": [98.0, 108.0], "gt_se_norm": [0.21212124824523926, 0.33580708503723145], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004823.pt"}
{"idx": 4825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdDX_uhxSwg.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman with blonde hair is closely examining the object in her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman with blonde hair is closely examining the object in her hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9257425665855408, -0.034653425216674805], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.18316829204559326, 0.03960394859313965], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004825.pt"}
{"idx": 4827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YdDX_uhxSwg.mp4", "duration": 26.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman in the middle pass an item to the young person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman in the middle pass an item to the young person?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [0.7079207897186279, 0.9306931495666504], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.7821782827377319, 0.9306931495666504], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004827.pt"}
{"idx": 4829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VJwEh7xne-g.mp4", "duration": 64.24011951650142, "prompt_text": "Please find the visual event described by the sentence 'When is the woman with glasses laughing heartily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman with glasses laughing heartily?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.813200831413269, -0.6264016628265381], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.1905369758605957, -0.097137451171875], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004829.pt"}
{"idx": 4831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VJwEh7xne-g.mp4", "duration": 64.24011951650142, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman in a green shirt walk by in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman in a green shirt walk by in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 37.0], "base_se_norm": [0.02739536762237549, 0.15192806720733643], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [0.02739536762237549, 0.08966171741485596], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004831.pt"}
{"idx": 4833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VJwEh7xne-g.mp4", "duration": 64.24011951650142, "prompt_text": "Please find the visual event described by the sentence 'Locate the instance where the woman places her hands flat on top of the man's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the instance where the woman places her hands flat on top of the man's hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 45.0], "base_se_norm": [-0.06600421667098999, 0.4009937047958374], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [0.8368583917617798, 0.9302579164505005], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004833.pt"}
{"idx": 4835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtBBL7Pb9Q0.mp4", "duration": 43.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the train is passing directly by the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the train is passing directly by the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 29.0], "base_se_norm": [-0.3098159432411194, 0.33435583114624023], "gt_se_raw": [16.0, 29.0], "gt_se_norm": [-0.2638036608695984, 0.33435583114624023], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004835.pt"}
{"idx": 4837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtBBL7Pb9Q0.mp4", "duration": 43.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman in a pink shirt crouches down as the train passes her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman in a pink shirt crouches down as the train passes her.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 29.0], "base_se_norm": [-0.17177915573120117, 0.33435583114624023], "gt_se_raw": [19.0, 24.0], "gt_se_norm": [-0.12576687335968018, 0.10429441928863525], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004837.pt"}
{"idx": 4839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lgbdir5RAHc.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'When is the woman alone, searching through her bag at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman alone, searching through her bag at the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 8.0], "base_se_norm": [-0.9896185398101807, -0.9169481992721558], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9896185398101807, -0.9169481992721558], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004839.pt"}
{"idx": 4841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lgbdir5RAHc.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman sits on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman sits on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 7.0], "base_se_norm": [-0.9896185398101807, -0.9273296594619751], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [-0.6158852577209473, -0.5951223373413086], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004841.pt"}
{"idx": 4843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lgbdir5RAHc.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'When do the two people hold hands before the man leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 143.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two people hold hands before the man leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 143.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 143.0], "base_se_norm": [0.4534069299697876, 0.48455142974853516], "gt_se_raw": [140.0, 142.0], "gt_se_norm": [0.4534069299697876, 0.4741699695587158], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004843.pt"}
{"idx": 4845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6NCN13K7Js.mp4", "duration": 81.01426666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when a boy is using picture cards to communicate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a boy is using picture cards to communicate.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.7531298995018005, -0.6296948194503784], "gt_se_raw": [10.0, 16.0], "gt_se_norm": [-0.7531298995018005, -0.6050078868865967], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004845.pt"}
{"idx": 4847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b6NCN13K7Js.mp4", "duration": 81.01426666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display pie charts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display pie charts?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 64.0], "base_se_norm": [0.38247251510620117, 0.5799685716629028], "gt_se_raw": [56.0, 60.0], "gt_se_norm": [0.38247251510620117, 0.4812206029891968], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004847.pt"}
{"idx": 4849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4mssLzXoSg.mp4", "duration": 141.76662499999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man getting out of a black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man getting out of a black car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 44.0], "base_se_norm": [-0.43569231033325195, -0.3792615532875061], "gt_se_raw": [40.0, 43.0], "gt_se_norm": [-0.43569231033325195, -0.39336925745010376], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004849.pt"}
{"idx": 4851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4mssLzXoSg.mp4", "duration": 141.76662499999998, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man in the beige coat start talking on his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man in the beige coat start talking on his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 67.0], "base_se_norm": [-0.18175387382507324, -0.05478465557098389], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.1958615779876709, -0.11121541261672974], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004851.pt"}
{"idx": 4853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4mssLzXoSg.mp4", "duration": 141.76662499999998, "prompt_text": "Please find the visual event described by the sentence 'A man runs frantically through traffic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man runs frantically through traffic.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 111.0], "base_se_norm": [0.4107692241668701, 0.5659538507461548], "gt_se_raw": [101.0, 106.0], "gt_se_norm": [0.4248769283294678, 0.4954153299331665], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004853.pt"}
{"idx": 4855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0brZ8Cinao.mp4", "duration": 29.766666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a girl with a ponytail writing in a notebook in a library?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a girl with a ponytail writing in a notebook in a library?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.32810747623443604, -0.12653976678848267], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.2609182596206665, -0.12653976678848267], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004855.pt"}
{"idx": 4857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/s0brZ8Cinao.mp4", "duration": 29.766666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a person with short red hair is smiling while using a computer in a dark room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a person with short red hair is smiling while using a computer in a dark room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [0.007838726043701172, 0.20940649509429932], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.07502806186676025, 0.20940649509429932], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004857.pt"}
{"idx": 4859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34aAAxaOABU.mp4", "duration": 69.43603333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the person is assembling the legs of the sculpture while kneeling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the person is assembling the legs of the sculpture while kneeling.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 18.0], "base_se_norm": [-0.9711965322494507, -0.48153722286224365], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9711965322494507, -0.5679476857185364], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004859.pt"}
{"idx": 4861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34aAAxaOABU.mp4", "duration": 69.43603333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the sculpture have a full body but no head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the sculpture have a full body but no head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 33.0], "base_se_norm": [-0.2799128293991089, -0.04948490858078003], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [0.20974647998809814, 0.32496047019958496], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004861.pt"}
{"idx": 4863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/34aAAxaOABU.mp4", "duration": 69.43603333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the completed sculpture being rotated and displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the completed sculpture being rotated and displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [0.814619779586792, 0.9586372375488281], "gt_se_raw": [64.0, 69.0], "gt_se_norm": [0.8434232473373413, 0.9874407052993774], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004863.pt"}
{"idx": 4865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SZAraIESRCY.mp4", "duration": 143.91043333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a group of people are shown using their smartphones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a group of people are shown using their smartphones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 33.0], "base_se_norm": [-0.6386641263961792, -0.5413813591003418], "gt_se_raw": [26.0, 38.0], "gt_se_norm": [-0.6386641263961792, -0.4718937277793884], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004865.pt"}
{"idx": 4867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SZAraIESRCY.mp4", "duration": 143.91043333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a foot in a black shoe steps on a yellow and red pedal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a foot in a black shoe steps on a yellow and red pedal.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [-0.096660315990448, -0.05496776103973389], "gt_se_raw": [66.0, 68.0], "gt_se_norm": [-0.08276277780532837, -0.05496776103973389], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004867.pt"}
{"idx": 4869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ADrkVCYoR2U.mp4", "duration": 43.00963333333333, "prompt_text": "Please find the visual event described by the sentence 'A baby is lying in a white round bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A baby is lying in a white round bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 30.0], "base_se_norm": [-1.0, 0.39503633975982666], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.906997561454773], "start_token_span": [70, 73], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004869.pt"}
{"idx": 4871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ADrkVCYoR2U.mp4", "duration": 43.00963333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the baby is being flipped upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the baby is being flipped upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 10.0], "base_se_norm": [-0.8139951229095459, -0.5349879264831543], "gt_se_raw": [6.0, 7.0], "gt_se_norm": [-0.7209927439689636, -0.6744915246963501], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004871.pt"}
{"idx": 4873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ADrkVCYoR2U.mp4", "duration": 43.00963333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the text 'Like: Spinning' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the text 'Like: Spinning' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 42.0], "base_se_norm": [0.5345399379730225, 0.9530508518218994], "gt_se_raw": [33.0, 43.0], "gt_se_norm": [0.5345399379730225, 0.9995521306991577], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004873.pt"}
{"idx": 4875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/peu8CIu1W8Q.mp4", "duration": 91.45803333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man with glasses and spiky hair introduce himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man with glasses and spiky hair introduce himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.7375845909118652, -0.6501127481460571], "gt_se_raw": [13.0, 16.0], "gt_se_norm": [-0.7157166004180908, -0.6501127481460571], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004875.pt"}
{"idx": 4877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/peu8CIu1W8Q.mp4", "duration": 91.45803333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when a man with a beard and glasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when a man with a beard and glasses is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 40.0], "base_se_norm": [-0.27835750579833984, -0.12528187036514282], "gt_se_raw": [33.0, 40.0], "gt_se_norm": [-0.27835750579833984, -0.12528187036514282], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004877.pt"}
{"idx": 4879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgZBmmuB_wg.mp4", "duration": 19.583333333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see a full view of the courtroom, including the judge, jury, and lawyer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a full view of the courtroom, including the judge, jury, and lawyer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.7957446575164795, -0.5914893746376038], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.8978723287582397, -0.5914893746376038], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004879.pt"}
{"idx": 4881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgZBmmuB_wg.mp4", "duration": 19.583333333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the TV-headed character look up with a worried expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the TV-headed character look up with a worried expression?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.4893617033958435, -0.18297874927520752], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [0.02127659320831299, 0.22553181648254395], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004881.pt"}
{"idx": 4883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgZBmmuB_wg.mp4", "duration": 19.583333333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'Just the facts' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'Just the facts' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [0.42978715896606445, 0.7361701726913452], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [0.5319148302078247, 0.7361701726913452], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004883.pt"}
{"idx": 4885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1PhPYr_9zRY.mp4", "duration": 233.35812499999997, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a puppy is licking a man's neck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a puppy is licking a man's neck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.9657179117202759, -0.9400063753128052], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9657179117202759, -0.9485768675804138], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004885.pt"}
{"idx": 4887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1PhPYr_9zRY.mp4", "duration": 233.35812499999997, "prompt_text": "Please find the visual event described by the sentence 'When does the puppy pee on the pink mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppy pee on the pink mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 85.0], "base_se_norm": [-0.28864699602127075, -0.2715059518814087], "gt_se_raw": [72.0, 75.0], "gt_se_norm": [-0.3829227089881897, -0.357211172580719], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004887.pt"}
{"idx": 4889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RqWYIB1O-0.mp4", "duration": 27.8, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the red shirt pick up the iPad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the red shirt pick up the iPad?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.6402877569198608, -0.3525179624557495], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.568345308303833, -0.42446041107177734], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004889.pt"}
{"idx": 4891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RqWYIB1O-0.mp4", "duration": 27.8, "prompt_text": "Please find the visual event described by the sentence 'At what point does the seated woman hand a piece of paper to the standing woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the seated woman hand a piece of paper to the standing woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [0.5827338695526123, 0.726618766784668], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.5827338695526123, 0.726618766784668], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004891.pt"}
{"idx": 4893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2RqWYIB1O-0.mp4", "duration": 27.8, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the woman at the desk is looking at her computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the woman at the desk is looking at her computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 22.0], "base_se_norm": [-1.0, 0.5827338695526123], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.8561151027679443, -0.6402877569198608], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004893.pt"}
{"idx": 4895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8fn4fewrvQg.mp4", "duration": 72.53912245578913, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the 'Environmental Health & Safety' webpage is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the 'Environmental Health & Safety' webpage is being shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.7794293761253357, -0.6691440343856812], "gt_se_raw": [7.0, 12.0], "gt_se_norm": [-0.8070006966590881, -0.6691440343856812], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004895.pt"}
{"idx": 4897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8fn4fewrvQg.mp4", "duration": 72.53912245578913, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment featuring the UWF Student, Virginia Morrison, speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment featuring the UWF Student, Virginia Morrison, speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 53.0], "base_se_norm": [0.26828110218048096, 0.4612804651260376], "gt_se_raw": [45.0, 53.0], "gt_se_norm": [0.24070978164672852, 0.4612804651260376], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004897.pt"}
{"idx": 4899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cB7O_JIxGuY.mp4", "duration": 126.4263, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the witch dips the apple into the bubbling green brew.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the witch dips the apple into the bubbling green brew.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.9367220401763916, -0.7943465709686279], "gt_se_raw": [4.0, 9.0], "gt_se_norm": [-0.9367220401763916, -0.8576245307922363], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004899.pt"}
{"idx": 4901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cB7O_JIxGuY.mp4", "duration": 126.4263, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the witch tempts the raven with the red apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the witch tempts the raven with the red apple.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 43.0], "base_se_norm": [-0.39885926246643066, -0.31976181268692017], "gt_se_raw": [38.0, 43.0], "gt_se_norm": [-0.39885926246643066, -0.31976181268692017], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004901.pt"}
{"idx": 4903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cB7O_JIxGuY.mp4", "duration": 126.4263, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the witch offers the apple to Snow White at the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the witch offers the apple to Snow White at the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 124.0], "base_se_norm": [0.6610467433929443, 0.961617112159729], "gt_se_raw": [117.0, 124.0], "gt_se_norm": [0.8508807420730591, 0.961617112159729], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004903.pt"}
{"idx": 4905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0JboM-STb4E.mp4", "duration": 151.23946248226358, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when two women are surprised by the appearance of a puppy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when two women are surprised by the appearance of a puppy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 35.0], "base_se_norm": [-0.6561744213104248, -0.5371577739715576], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.6429502964019775, -0.6165021657943726], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004905.pt"}
{"idx": 4907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0JboM-STb4E.mp4", "duration": 151.23946248226358, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman with braids is crying with a big smile.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman with braids is crying with a big smile.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 55.0], "base_se_norm": [-0.29912471771240234, -0.27267658710479736], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [-0.29912471771240234, -0.27267658710479736], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004907.pt"}
{"idx": 4909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24Pmmo9wKik.mp4", "duration": 70.87100000000001, "prompt_text": "Please find the visual event described by the sentence 'When does the cartoon character type on the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cartoon character type on the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 14.0], "base_se_norm": [-0.8588985800743103, -0.604915976524353], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.8306782841682434, -0.7460174560546875], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004909.pt"}
{"idx": 4911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24Pmmo9wKik.mp4", "duration": 70.87100000000001, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man yells 'Lasse' for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man yells 'Lasse' for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 39.0], "base_se_norm": [0.01593029499053955, 0.10059118270874023], "gt_se_raw": [37.0, 39.0], "gt_se_norm": [0.044150590896606445, 0.10059118270874023], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004911.pt"}
{"idx": 4913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/24Pmmo9wKik.mp4", "duration": 70.87100000000001, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a woman turns on the lights with a remote control.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a woman turns on the lights with a remote control.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [0.8343186378479004, 0.9189795255661011], "gt_se_raw": [64.0, 67.0], "gt_se_norm": [0.8060983419418335, 0.8907592296600342], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004913.pt"}
{"idx": 4915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNi_HC839Wo.mp4", "duration": 153.06957777777777, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the girl clarifies why she calls herself 'Lady Bird'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the girl clarifies why she calls herself 'Lady Bird'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.660285234451294, -0.6080213785171509], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.6472193002700806, -0.6080213785171509], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004915.pt"}
{"idx": 4917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNi_HC839Wo.mp4", "duration": 153.06957777777777, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the scene where the girl and the boy are getting intimate in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the scene where the girl and the boy are getting intimate in the bedroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 89.0], "base_se_norm": [0.11060607433319092, 0.16286981105804443], "gt_se_raw": [85.0, 92.0], "gt_se_norm": [0.11060607433319092, 0.20206773281097412], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004917.pt"}
{"idx": 4919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hkmvuV6PK20.mp4", "duration": 162.82933333333332, "prompt_text": "Please find the visual event described by the sentence 'When is the boy crawling on the floor and crying for his mother?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy crawling on the floor and crying for his mother?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 33.0], "base_se_norm": [-0.8403236269950867, -0.5946676135063171], "gt_se_raw": [8.0, 16.0], "gt_se_norm": [-0.901737630367279, -0.8034752011299133], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004919.pt"}
{"idx": 4921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hkmvuV6PK20.mp4", "duration": 162.82933333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy is warming his hands by the fireplace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy is warming his hands by the fireplace.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [-0.07879006862640381, -0.017376065254211426], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.05422443151473999, -0.005093276500701904], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004921.pt"}
{"idx": 4923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hkmvuV6PK20.mp4", "duration": 162.82933333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the mother embrace her son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the mother embrace her son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 160.0], "base_se_norm": [0.866985559463501, 0.9652478694915771], "gt_se_raw": [152.0, 158.0], "gt_se_norm": [0.866985559463501, 0.9406822919845581], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004923.pt"}
{"idx": 4925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZEnkdcZuhnQ.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does she talk about her job title?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she talk about her job title?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9867403507232666, -0.8674033284187317], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.7613259553909302, -0.7215469479560852], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004925.pt"}
{"idx": 4927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZEnkdcZuhnQ.mp4", "duration": 150.83333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does she mention the ITAC selection center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does she mention the ITAC selection center?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.8674033284187317, -0.8143646717071533], "gt_se_raw": [48.0, 51.0], "gt_se_norm": [-0.36353588104248047, -0.3237568736076355], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004927.pt"}
{"idx": 4929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fj3u4Wd2C-8.mp4", "duration": 155.48866666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the text 'HOW DOES IT WORK AND HOW DOES IT HELP?' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the text 'HOW DOES IT WORK AND HOW DOES IT HELP?' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 69.0], "base_se_norm": [-0.1382008194923401, -0.11247551441192627], "gt_se_raw": [67.0, 69.0], "gt_se_norm": [-0.1382008194923401, -0.11247551441192627], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004929.pt"}
{"idx": 4931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fj3u4Wd2C-8.mp4", "duration": 155.48866666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the time when a woman with a disability is demonstrated using the eye-gaze technology.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time when a woman with a disability is demonstrated using the eye-gaze technology.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 121.0], "base_se_norm": [0.46634483337402344, 0.5563836097717285], "gt_se_raw": [115.0, 122.0], "gt_se_norm": [0.47920751571655273, 0.5692462921142578], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004931.pt"}
{"idx": 4933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fj3u4Wd2C-8.mp4", "duration": 155.48866666666666, "prompt_text": "Please find the visual event described by the sentence 'When is an image of a university building shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is an image of a university building shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 148.0], "base_se_norm": [0.8650877475738525, 0.9036757946014404], "gt_se_raw": [146.0, 148.0], "gt_se_norm": [0.8779504299163818, 0.9036757946014404], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004933.pt"}
{"idx": 4935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F-eMt3SrfFU.mp4", "duration": 138.7636222222222, "prompt_text": "Please find the visual event described by the sentence 'At what point are soldiers seen scrambling to climb onto a large white ship?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are soldiers seen scrambling to climb onto a large white ship?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 63.0], "base_se_norm": [-0.13522005081176758, -0.09198105335235596], "gt_se_raw": [61.0, 63.0], "gt_se_norm": [-0.12080705165863037, -0.09198105335235596], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004935.pt"}
{"idx": 4937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F-eMt3SrfFU.mp4", "duration": 138.7636222222222, "prompt_text": "Please find the visual event described by the sentence 'When can we see three fighter planes flying in formation in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three fighter planes flying in formation in the sky?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [-0.09198105335235596, -0.048742055892944336], "gt_se_raw": [96.0, 98.0], "gt_se_norm": [0.3836479187011719, 0.4124739170074463], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004937.pt"}
{"idx": 4939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCdx7-XOmVc.mp4", "duration": 222.48893333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the pictures of the six world leaders shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the pictures of the six world leaders shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 43.0], "base_se_norm": [-0.757291316986084, -0.6134639382362366], "gt_se_raw": [26.0, 43.0], "gt_se_norm": [-0.7662805318832397, -0.6134639382362366], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004939.pt"}
{"idx": 4941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCdx7-XOmVc.mp4", "duration": 222.48893333333334, "prompt_text": "Please find the visual event described by the sentence 'Can you show me the moment when the term 'GDP' is explained as 'Gross Domestic Product'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Can you show me the moment when the term 'GDP' is explained as 'Gross Domestic Product'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.6044747233390808, -0.5685178637504578], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [-0.595485508441925, -0.5685178637504578], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004941.pt"}
{"idx": 4943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCdx7-XOmVc.mp4", "duration": 222.48893333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the phrase 'World Economic Systems' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 206.0 - 209.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the phrase 'World Economic Systems' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 206.0 - 209.0 seconds.<|im_end|>\n", "base_se_raw": [206.0, 209.0], "base_se_norm": [0.8517774343490601, 0.8787450790405273], "gt_se_raw": [207.0, 211.0], "gt_se_norm": [0.8607666492462158, 0.8967235088348389], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004943.pt"}
{"idx": 4945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NVGGgncVq-4.mp4", "duration": 253.24, "prompt_text": "Please find the visual event described by the sentence 'At what point is the girl watching news about natural disasters on TV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the girl watching news about natural disasters on TV?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 49.0], "base_se_norm": [-0.8183541297912598, -0.6130152940750122], "gt_se_raw": [24.0, 40.0], "gt_se_norm": [-0.8104565143585205, -0.6840941905975342], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004945.pt"}
{"idx": 4947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NVGGgncVq-4.mp4", "duration": 253.24, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the girl is running through the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the girl is running through the desert.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 116.0], "base_se_norm": [-0.12336122989654541, -0.08387303352355957], "gt_se_raw": [110.0, 123.0], "gt_se_norm": [-0.13125890493392944, -0.028589487075805664], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004947.pt"}
{"idx": 4949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mgNgscHJh6I.mp4", "duration": 67.06700000000001, "prompt_text": "Please find the visual event described by the sentence 'When can we see the blue character falling into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the blue character falling into the manhole?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.6719698309898376, -0.6123279929161072], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.7316116690635681, -0.6421489119529724], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004949.pt"}
{"idx": 4951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mgNgscHJh6I.mp4", "duration": 67.06700000000001, "prompt_text": "Please find the visual event described by the sentence 'At what point does a bald man approach the two women and touch one on the shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a bald man approach the two women and touch one on the shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 38.0], "base_se_norm": [-0.10537225008010864, 0.133195161819458], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.10537225008010864, 0.01391148567199707], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004951.pt"}
{"idx": 4953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mgNgscHJh6I.mp4", "duration": 67.06700000000001, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the letter's content is fully displayed and read aloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the letter's content is fully displayed and read aloud.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 53.0], "base_se_norm": [0.34194159507751465, 0.5805090665817261], "gt_se_raw": [56.0, 63.0], "gt_se_norm": [0.6699718236923218, 0.8787182569503784], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004953.pt"}
{"idx": 4955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdEfx8up7wM.mp4", "duration": 68.44337499999999, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the man in the red plaid shirt makes a long farting sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the man in the red plaid shirt makes a long farting sound.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [-0.09414166212081909, -0.006477952003479004], "gt_se_raw": [11.0, 12.0], "gt_se_norm": [-0.6785663962364197, -0.6493451595306396], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004955.pt"}
{"idx": 4957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OdEfx8up7wM.mp4", "duration": 68.44337499999999, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the boy in the blue shirt smells the can and faints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the boy in the blue shirt smells the can and faints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 48.0], "base_se_norm": [0.19807064533233643, 0.4026193618774414], "gt_se_raw": [46.0, 48.0], "gt_se_norm": [0.34417688846588135, 0.4026193618774414], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004957.pt"}
{"idx": 4959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvHSlHhh1gk.mp4", "duration": 144.14399999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the man in the orange vest walking with a coffee cup and greeting people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the man in the orange vest walking with a coffee cup and greeting people?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 36.0], "base_se_norm": [-0.6114996075630188, -0.5004994869232178], "gt_se_raw": [28.0, 36.0], "gt_se_norm": [-0.6114996075630188, -0.5004994869232178], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004959.pt"}
{"idx": 4961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvHSlHhh1gk.mp4", "duration": 144.14399999999998, "prompt_text": "Please find the visual event described by the sentence 'When does the flying vehicle shoot pink hearts at the man running on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the flying vehicle shoot pink hearts at the man running on the bridge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.08424907922744751, -0.014873981475830078], "gt_se_raw": [74.0, 75.0], "gt_se_norm": [0.026751041412353516, 0.04062604904174805], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004961.pt"}
{"idx": 4963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XvHSlHhh1gk.mp4", "duration": 144.14399999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the shot where the spaceship is flying through a galaxy of colorful planets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the shot where the spaceship is flying through a galaxy of colorful planets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 132.0], "base_se_norm": [0.7760018110275269, 0.831501841545105], "gt_se_raw": [128.0, 132.0], "gt_se_norm": [0.7760018110275269, 0.831501841545105], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004963.pt"}
{"idx": 4965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfQHp_px7iI.mp4", "duration": 32.16, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a hand attaches the hose to the vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a hand attaches the hose to the vacuum cleaner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 11.0], "base_se_norm": [-0.5024875402450562, -0.3159204125404358], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.44029849767684937, -0.3159204125404358], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004965.pt"}
{"idx": 4967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hfQHp_px7iI.mp4", "duration": 32.16, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a person is actively vacuuming the rug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a person is actively vacuuming the rug.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.06716418266296387, 0.3681591749191284], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.06716418266296387, 0.3681591749191284], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004967.pt"}
{"idx": 4969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FuraQCCsKgE.mp4", "duration": 8.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dog walking towards the staircase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dog walking towards the staircase?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.5256917476654053], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.7628458738327026], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004969.pt"}
{"idx": 4971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FuraQCCsKgE.mp4", "duration": 8.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the dog drop the toy and look up before vocalizing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog drop the toy and look up before vocalizing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.28853756189346313, 0.18577075004577637], "gt_se_raw": [4.0, 5.0], "gt_se_norm": [-0.05138343572616577, 0.18577075004577637], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004971.pt"}
{"idx": 4973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FuraQCCsKgE.mp4", "duration": 8.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'Isolate the moment the dog turns from the wall back towards the center of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Isolate the moment the dog turns from the wall back towards the center of the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.05138343572616577, 0.42292487621307373], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.7628458738327026, -0.5256917476654053], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004973.pt"}
{"idx": 4975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YEG08_h6aT8.mp4", "duration": 150.6, "prompt_text": "Please find the visual event described by the sentence 'Show the raccoon eating red berries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the raccoon eating red berries.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 53.0], "base_se_norm": [-0.5219123363494873, -0.29614877700805664], "gt_se_raw": [36.0, 53.0], "gt_se_norm": [-0.5219123363494873, -0.29614877700805664], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004975.pt"}
{"idx": 4977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YEG08_h6aT8.mp4", "duration": 150.6, "prompt_text": "Please find the visual event described by the sentence 'Find the part of the video that introduces fish swimming in the river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part of the video that introduces fish swimming in the river.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 95.0], "base_se_norm": [0.11553776264190674, 0.2616201639175415], "gt_se_raw": [85.0, 96.0], "gt_se_norm": [0.12881803512573242, 0.27490031719207764], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004977.pt"}
{"idx": 4979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2y1QQWNZxZM.mp4", "duration": 33.06636666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the man holding the child in his arms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man holding the child in his arms?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.8185467720031738], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.8185467720031738], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004979.pt"}
{"idx": 4981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2y1QQWNZxZM.mp4", "duration": 33.06636666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child pull themselves up using the blue couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child pull themselves up using the blue couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [0.39114153385162354, 0.69356369972229], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [0.5121104717254639, 0.69356369972229], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004981.pt"}
{"idx": 4983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2y1QQWNZxZM.mp4", "duration": 33.06636666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man position the child on the floor to start crawling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man position the child on the floor to start crawling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 10.0], "base_se_norm": [-0.8790311813354492, -0.3951558470726013], "gt_se_raw": [3.0, 9.0], "gt_se_norm": [-0.8185467720031738, -0.4556402564048767], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004983.pt"}
{"idx": 4985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uy7ykWjFH1E.mp4", "duration": 143.6435, "prompt_text": "Please find the visual event described by the sentence 'Show the slide that explains how smoke alarms for deaf people work, featuring the NFPA logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the slide that explains how smoke alarms for deaf people work, featuring the NFPA logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 51.0], "base_se_norm": [-0.6519160270690918, -0.28990864753723145], "gt_se_raw": [26.0, 51.0], "gt_se_norm": [-0.6379926204681396, -0.28990864753723145], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004985.pt"}
{"idx": 4987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uy7ykWjFH1E.mp4", "duration": 143.6435, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the acronym 'CLIS' and its details are displayed on a white background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the acronym 'CLIS' and its details are displayed on a white background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 116.0], "base_se_norm": [0.1277921199798584, 0.6151096820831299], "gt_se_raw": [82.0, 116.0], "gt_se_norm": [0.14171552658081055, 0.6151096820831299], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004987.pt"}
{"idx": 4989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wji-BZ0oCwg.mp4", "duration": 117.32554166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a person in a white cape is walking on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a person in a white cape is walking on the beach.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 78.0], "base_se_norm": [0.29554057121276855, 0.32963383197784424], "gt_se_raw": [77.0, 78.0], "gt_se_norm": [0.3125871419906616, 0.32963383197784424], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004989.pt"}
{"idx": 4991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wji-BZ0oCwg.mp4", "duration": 117.32554166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman in the grey scarf fights stormtroopers, leading to an explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman in the grey scarf fights stormtroopers, leading to an explosion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [-0.5226955413818359, -0.4545091986656189], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.5226955413818359, -0.4545091986656189], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004991.pt"}
{"idx": 4993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wji-BZ0oCwg.mp4", "duration": 117.32554166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the time range showing soldiers running on a beach during a battle with giant four-legged walkers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time range showing soldiers running on a beach during a battle with giant four-legged walkers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 84.0], "base_se_norm": [0.36372697353363037, 0.43191325664520264], "gt_se_raw": [81.0, 85.0], "gt_se_norm": [0.38077354431152344, 0.44895994663238525], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004993.pt"}
{"idx": 4995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KJiCalhfN6k.mp4", "duration": 22.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the person points at the 'weekly review' card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the person points at the 'weekly review' card.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.9113737344741821, -0.7341211438179016], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9113737344741821, -0.7341211438179016], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004995.pt"}
{"idx": 4997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KJiCalhfN6k.mp4", "duration": 22.566666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the individual is interacting with the schedule board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the individual is interacting with the schedule board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 22.0], "base_se_norm": [-0.9113737344741821, 0.9497783184051514], "gt_se_raw": [0.0, 18.0], "gt_se_norm": [-1.0, 0.5952732563018799], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004997.pt"}
{"idx": 4999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/znxRGH92XDg.mp4", "duration": 157.6575, "prompt_text": "Please find the visual event described by the sentence 'When are the students seen exiting the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the students seen exiting the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.9111999273300171], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.9365713596343994], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00004999.pt"}
{"idx": 5001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/znxRGH92XDg.mp4", "duration": 157.6575, "prompt_text": "Please find the visual event described by the sentence 'At what point do the woman and man engage in a heated discussion on the bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the woman and man engage in a heated discussion on the bus?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 69.0], "base_se_norm": [-0.3657136559486389, -0.12468487024307251], "gt_se_raw": [50.0, 64.0], "gt_se_norm": [-0.3657136559486389, -0.18811345100402832], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005001.pt"}
{"idx": 5003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/znxRGH92XDg.mp4", "duration": 157.6575, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the bus collides with the trolley.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the bus collides with the trolley.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 82.0], "base_se_norm": [-0.0105133056640625, 0.04022955894470215], "gt_se_raw": [88.0, 95.0], "gt_se_norm": [0.11634397506713867, 0.20514404773712158], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005003.pt"}
{"idx": 5005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ate2ecHwBoo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the time range where a group of people are watching a man lead a brown horse outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range where a group of people are watching a man lead a brown horse outdoors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5409404635429382, -0.08188092708587646], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005005.pt"}
{"idx": 5007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ate2ecHwBoo.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a smiling woman stands with a horse, and the text 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a smiling woman stands with a horse, and the text 'YOU'VE GOT THIS!' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005007.pt"}
{"idx": 5009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2qaNcyIoLV8.mp4", "duration": 132.76596666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see the Australian Coat of Arms on the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the Australian Coat of Arms on the building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.8041666746139526, -0.7589743733406067], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.8041666746139526, -0.7740384340286255], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005009.pt"}
{"idx": 5011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2qaNcyIoLV8.mp4", "duration": 132.76596666666666, "prompt_text": "Please find the visual event described by the sentence 'Where are the empty red seats of the Senate chamber shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where are the empty red seats of the Senate chamber shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 123.0], "base_se_norm": [0.6118589639663696, 0.8528845310211182], "gt_se_raw": [107.0, 123.0], "gt_se_norm": [0.6118589639663696, 0.8528845310211182], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005011.pt"}
{"idx": 5013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2qaNcyIoLV8.mp4", "duration": 132.76596666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene with the restroom signs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene with the restroom signs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 106.0], "base_se_norm": [0.476282000541687, 0.596794843673706], "gt_se_raw": [97.0, 107.0], "gt_se_norm": [0.46121788024902344, 0.6118589639663696], "start_token_span": [63, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005013.pt"}
{"idx": 5015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R_j_5joOBL0.mp4", "duration": 126.7266, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman drinks water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman drinks water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 71.0], "base_se_norm": [-0.005733609199523926, 0.12052249908447266], "gt_se_raw": [62.0, 69.0], "gt_se_norm": [-0.021515607833862305, 0.08895838260650635], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005015.pt"}
{"idx": 5017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R_j_5joOBL0.mp4", "duration": 126.7266, "prompt_text": "Please find the visual event described by the sentence 'When does the person in the white sweater bring a drink to the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in the white sweater bring a drink to the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9526540040969849, -0.8737439513206482], "gt_se_raw": [7.0, 10.0], "gt_se_norm": [-0.8895259499549866, -0.8421799540519714], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005017.pt"}
{"idx": 5019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aI51UWF8_9Y.mp4", "duration": 115.98253333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman doing a lower body workout in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman doing a lower body workout in a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 34.0], "base_se_norm": [-0.4826807379722595, -0.4137048125267029], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.4826807379722595, -0.4137048125267029], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005019.pt"}
{"idx": 5021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aI51UWF8_9Y.mp4", "duration": 115.98253333333334, "prompt_text": "Please find the visual event described by the sentence 'A man is lying on the floor and lifting a child as a weight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is lying on the floor and lifting a child as a weight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [0.17259037494659424, 0.22432219982147217], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [0.17259037494659424, 0.22432219982147217], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005021.pt"}
{"idx": 5023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aI51UWF8_9Y.mp4", "duration": 115.98253333333334, "prompt_text": "Please find the visual event described by the sentence 'A woman is practicing soccer skills by herself on a grassy field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A woman is practicing soccer skills by herself on a grassy field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 75.0], "base_se_norm": [0.24156618118286133, 0.2932981252670288], "gt_se_raw": [90.0, 95.0], "gt_se_norm": [0.5519577264785767, 0.6381776332855225], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005023.pt"}
{"idx": 5025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6pBONClIuzg.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is lying on a hospital bed and screaming in pain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is lying on a hospital bed and screaming in pain.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 81.0], "base_se_norm": [0.13094449043273926, 0.4313516616821289], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [0.37833869457244873, 0.4313516616821289], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005025.pt"}
{"idx": 5027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6pBONClIuzg.mp4", "duration": 113.17973333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when we can see a crash site at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when we can see a crash site at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 60.0], "base_se_norm": [0.024918437004089355, 0.06026053428649902], "gt_se_raw": [58.0, 60.0], "gt_se_norm": [0.024918437004089355, 0.06026053428649902], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005027.pt"}
{"idx": 5029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ck2K69zZBG4.mp4", "duration": 136.42795833333332, "prompt_text": "Please find the visual event described by the sentence 'At what time can we see only the plants against the pink and blue background at the start of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time can we see only the plants against the pink and blue background at the start of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 8.0], "base_se_norm": [-1.0, -0.882722020149231], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.882722020149231], "start_token_span": [76, 79], "end_token_span": [81, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005029.pt"}
{"idx": 5031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ck2K69zZBG4.mp4", "duration": 136.42795833333332, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the man is pointing at the plant on his right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the man is pointing at the plant on his right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 75.0], "base_se_norm": [-0.20837342739105225, 0.09948134422302246], "gt_se_raw": [62.0, 75.0], "gt_se_norm": [-0.0910954475402832, 0.09948134422302246], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005031.pt"}
{"idx": 5033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ck2K69zZBG4.mp4", "duration": 136.42795833333332, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the text 'SUBSCRIBE TO DAD' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the text 'SUBSCRIBE TO DAD' is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 136.0], "base_se_norm": [0.7151908874511719, 0.9937261343002319], "gt_se_raw": [117.0, 136.0], "gt_se_norm": [0.7151908874511719, 0.9937261343002319], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005033.pt"}
{"idx": 5035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EthCVn45VyU.mp4", "duration": 45.8, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the person is drying their hands with a paper towel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the person is drying their hands with a paper towel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 44.0], "base_se_norm": [0.5283843278884888, 0.9213974475860596], "gt_se_raw": [37.0, 44.0], "gt_se_norm": [0.6157205104827881, 0.9213974475860596], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005035.pt"}
{"idx": 5037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EthCVn45VyU.mp4", "duration": 45.8, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the person creates a soapy lather.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the person creates a soapy lather.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 26.0], "base_se_norm": [-0.6506550312042236, 0.13537120819091797], "gt_se_raw": [8.0, 27.0], "gt_se_norm": [-0.6506550312042236, 0.1790393590927124], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005037.pt"}
{"idx": 5039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVzDVbkDleE.mp4", "duration": 25.158466666666666, "prompt_text": "Please find the visual event described by the sentence 'When does SpongeBob realize that something is wrong with his brain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does SpongeBob realize that something is wrong with his brain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.682015597820282], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9205039143562317, -0.7615116834640503], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005039.pt"}
{"idx": 5041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVzDVbkDleE.mp4", "duration": 25.158466666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does Plankton start talking from inside SpongeBob's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Plankton start talking from inside SpongeBob's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 15.0], "base_se_norm": [-0.6025195121765137, 0.19244158267974854], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.28453510999679565, 0.19244158267974854], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005041.pt"}
{"idx": 5043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVzDVbkDleE.mp4", "duration": 25.158466666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where a possessed SpongeBob laughs evilly and walks out of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where a possessed SpongeBob laughs evilly and walks out of the frame.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 22.0], "base_se_norm": [0.27193760871887207, 0.7489142417907715], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [0.5104259252548218, 0.7489142417907715], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005043.pt"}
{"idx": 5045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ajzArdLK6tE.mp4", "duration": 143.56008333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the dill pickles are being labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the dill pickles are being labeled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 42.0], "base_se_norm": [-0.48453640937805176, -0.4148791432380676], "gt_se_raw": [37.0, 41.0], "gt_se_norm": [-0.48453640937805176, -0.42881059646606445], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005045.pt"}
{"idx": 5047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ajzArdLK6tE.mp4", "duration": 143.56008333333332, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the cucumbers and red chili are placed into the jar for the bread and butter pickles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the cucumbers and red chili are placed into the jar for the bread and butter pickles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 76.0], "base_se_norm": [-0.024798572063446045, 0.05879008769989014], "gt_se_raw": [70.0, 74.0], "gt_se_norm": [-0.024798572063446045, 0.030927181243896484], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005047.pt"}
{"idx": 5049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksqJTtIDgpw.mp4", "duration": 56.08, "prompt_text": "Please find the visual event described by the sentence 'When can we see a sign with 'HIGH VOLTAGE' written on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a sign with 'HIGH VOLTAGE' written on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 4.0], "base_se_norm": [-0.9286733269691467, -0.8573466539382935], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.964336633682251, -0.8930100202560425], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005049.pt"}
{"idx": 5051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksqJTtIDgpw.mp4", "duration": 56.08, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman is picking up her bow and arrows from the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman is picking up her bow and arrows from the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [-0.35805994272232056, -0.215406596660614], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.35805994272232056, -0.25106990337371826], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005051.pt"}
{"idx": 5053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ksqJTtIDgpw.mp4", "duration": 56.08, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of the deer's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of the deer's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 55.0], "base_se_norm": [0.9258202314376831, 0.9614834785461426], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [0.9258202314376831, 0.9971468448638916], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005053.pt"}
{"idx": 5055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I1Ksh6qrFNc.mp4", "duration": 29.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man is singing and dancing next to the silver car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man is singing and dancing next to the silver car.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.1314030885696411], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.2650333642959595, -0.06458795070648193], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005055.pt"}
{"idx": 5057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I1Ksh6qrFNc.mp4", "duration": 29.933333333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man makes a gesture like he is tasting something delicious.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man makes a gesture like he is tasting something delicious.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.06458795070648193, 0.06904232501983643], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.06458795070648193, 0.06904232501983643], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005057.pt"}
{"idx": 5059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i8Ol7OPpnbo.mp4", "duration": 27.260566666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we hear the phrase 'You're very crazy today'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we hear the phrase 'You're very crazy today'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.11960744857788086, 0.027124643325805664], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.4130716323852539, -0.2663395404815674], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005059.pt"}
{"idx": 5061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i8Ol7OPpnbo.mp4", "duration": 27.260566666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time does one of the girls ask 'Did you have fun on Parrot week kids?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does one of the girls ask 'Did you have fun on Parrot week kids?'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9266339540481567, -0.6331697702407837], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [0.1738567352294922, 0.3205888271331787], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005061.pt"}
{"idx": 5063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i8Ol7OPpnbo.mp4", "duration": 27.260566666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the girl in black picks up her napkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the girl in black picks up her napkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.687419056892395, 0.8341511487960815], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [0.8341511487960815, 0.9808832406997681], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005063.pt"}
{"idx": 5065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkX3eBvPDFQ.mp4", "duration": 301.88, "prompt_text": "Please find the visual event described by the sentence 'At what point does the ostrich get a watermelon rind stuck on its head, resembling a helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the ostrich get a watermelon rind stuck on its head, resembling a helmet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 167.0], "base_se_norm": [0.02689814567565918, 0.10639989376068115], "gt_se_raw": [154.0, 157.0], "gt_se_norm": [0.02027297019958496, 0.040148377418518066], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005065.pt"}
{"idx": 5067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KkX3eBvPDFQ.mp4", "duration": 301.88, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the ostrich falls into a giant spiderweb.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 215.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the ostrich falls into a giant spiderweb.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 215.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [215.0, 222.0], "base_se_norm": [0.4244070053100586, 0.4707831144332886], "gt_se_raw": [214.0, 217.0], "gt_se_norm": [0.4177818298339844, 0.43765735626220703], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005067.pt"}
{"idx": 5069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h3gV_MPxr3I.mp4", "duration": 153.01953333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman demonstrate how to lie on the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman demonstrate how to lie on the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 104.0], "base_se_norm": [0.17632043361663818, 0.3593035936355591], "gt_se_raw": [58.0, 62.0], "gt_se_norm": [-0.24192684888839722, -0.1896459460258484], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005069.pt"}
{"idx": 5071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h3gV_MPxr3I.mp4", "duration": 153.01953333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when all three individuals are interacting with the ball simultaneously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when all three individuals are interacting with the ball simultaneously.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 52.0], "base_se_norm": [-0.49026113748550415, -0.32034820318222046], "gt_se_raw": [69.0, 71.0], "gt_se_norm": [-0.09815436601638794, -0.07201391458511353], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005071.pt"}
{"idx": 5073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/h3gV_MPxr3I.mp4", "duration": 153.01953333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the woman and the boy pat the top of the ball together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the woman and the boy pat the top of the ball together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 52.0], "base_se_norm": [-0.4379802346229553, -0.32034820318222046], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.8562275171279907, -0.7778061628341675], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005073.pt"}
{"idx": 5075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ha92yzmVv2s.mp4", "duration": 113.7, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the black dog is next to the other dog on the deck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the black dog is next to the other dog on the deck.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9824098348617554, -0.9120492339134216], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9824098348617554, -0.9296393990516663], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005075.pt"}
{"idx": 5077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ha92yzmVv2s.mp4", "duration": 113.7, "prompt_text": "Please find the visual event described by the sentence 'When does the person throw the toy for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person throw the toy for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.8065083622932434, -0.718557596206665], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.7361477613449097, -0.7009674310684204], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005077.pt"}
{"idx": 5079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g6KUxEfUVm0.mp4", "duration": 49.816433333333336, "prompt_text": "Please find the visual event described by the sentence 'When is a woman massaging another woman's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman massaging another woman's arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 12.0], "base_se_norm": [-0.7591156363487244, -0.5182312726974487], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.7591156363487244, -0.5182312726974487], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005079.pt"}
{"idx": 5081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g6KUxEfUVm0.mp4", "duration": 49.816433333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman receiving a back massage while seated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman receiving a back massage while seated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.7992630004882812, -0.5182312726974487], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.116757333278656, 0.12412703037261963], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005081.pt"}
{"idx": 5083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g6KUxEfUVm0.mp4", "duration": 49.816433333333336, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman receives a neck massage and a hug from behind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman receives a neck massage and a hug from behind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 48.0], "base_se_norm": [0.48545360565185547, 0.9270749092102051], "gt_se_raw": [37.0, 48.0], "gt_se_norm": [0.48545360565185547, 0.9270749092102051], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005083.pt"}
{"idx": 5085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yZyxJxR6RCA.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a woman with blonde hair is editing on a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a woman with blonde hair is editing on a computer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.47593390941619873, -0.08288431167602539], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.47593390941619873, -0.08288431167602539], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005085.pt"}
{"idx": 5087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yZyxJxR6RCA.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a man in a black shirt stands with his arms crossed, smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a man in a black shirt stands with his arms crossed, smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.4411817789077759, 0.8342313766479492], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.4411817789077759, 0.8342313766479492], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005087.pt"}
{"idx": 5089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZ0Bey4YUGI.mp4", "duration": 142.64249999999998, "prompt_text": "Please find the visual event described by the sentence 'When is the movie title, The BFG, shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the movie title, The BFG, shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 126.0], "base_se_norm": [0.682528018951416, 0.766654372215271], "gt_se_raw": [120.0, 125.0], "gt_se_norm": [0.682528018951416, 0.7526333332061768], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005089.pt"}
{"idx": 5091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZ0Bey4YUGI.mp4", "duration": 142.64249999999998, "prompt_text": "Please find the visual event described by the sentence 'A girl is placed inside a cooking pot by a giant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A girl is placed inside a cooking pot by a giant.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [0.07962214946746826, 0.13570642471313477], "gt_se_raw": [76.0, 79.0], "gt_se_norm": [0.06560111045837402, 0.10766422748565674], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005091.pt"}
{"idx": 5093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZ0Bey4YUGI.mp4", "duration": 142.64249999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a cat sleeping on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a cat sleeping on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.5513259172439575, -0.5092626810073853], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.5513259172439575, -0.5232837200164795], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005093.pt"}
{"idx": 5095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a7CpzJ-sNl8.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a cargo ship sailing on the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a cargo ship sailing on the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 34.0], "base_se_norm": [0.03301835060119629, 0.13298797607421875], "gt_se_raw": [31.0, 34.0], "gt_se_norm": [0.03301835060119629, 0.13298797607421875], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005095.pt"}
{"idx": 5097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a7CpzJ-sNl8.mp4", "duration": 60.01829166666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a forklift is in operation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a forklift is in operation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 30.0], "base_se_norm": [-0.03362792730331421, -0.00030475854873657227], "gt_se_raw": [29.0, 31.0], "gt_se_norm": [-0.03362792730331421, 0.03301835060119629], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005097.pt"}
{"idx": 5099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cqDG7q508iM.mp4", "duration": 330.5969333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the middle explain that there are twelve eggs, with eight being hard-boiled and four being raw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the middle explain that there are twelve eggs, with eight being hard-boiled and four being raw?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 52.0], "base_se_norm": [-0.7822121381759644, -0.6854175329208374], "gt_se_raw": [17.0, 24.0], "gt_se_norm": [-0.8971557021141052, -0.8548080921173096], "start_token_span": [79, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005099.pt"}
{"idx": 5101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cqDG7q508iM.mp4", "duration": 330.5969333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the host smash a raw egg on his head for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the host smash a raw egg on his head for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 109.0], "base_se_norm": [-0.42528200149536133, -0.34058672189712524], "gt_se_raw": [108.0, 111.0], "gt_se_norm": [-0.34663641452789307, -0.3284873962402344], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005101.pt"}
{"idx": 5103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cqDG7q508iM.mp4", "duration": 330.5969333333333, "prompt_text": "Please find the visual event described by the sentence 'When do both men smash an egg on their heads simultaneously to decide the winner?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 286.0 - 292.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do both men smash an egg on their heads simultaneously to decide the winner?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 286.0 - 292.0 seconds.<|im_end|>\n", "base_se_raw": [286.0, 292.0], "base_se_norm": [0.7302036285400391, 0.7665016651153564], "gt_se_raw": [286.0, 289.0], "gt_se_norm": [0.7302036285400391, 0.7483526468276978], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005103.pt"}
{"idx": 5105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hlgcYUM35yY.mp4", "duration": 177.16, "prompt_text": "Please find the visual event described by the sentence 'When does she tie half of her hair into a bun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does she tie half of her hair into a bun?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.6387446522712708, -0.5935877561569214], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [-0.6274554133415222, -0.5822985172271729], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005105.pt"}
{"idx": 5107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hlgcYUM35yY.mp4", "duration": 177.16, "prompt_text": "Please find the visual event described by the sentence 'At what point does she start braiding the second pigtail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 161.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she start braiding the second pigtail?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 161.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 161.0], "base_se_norm": [0.6369383335113525, 0.8175660371780396], "gt_se_raw": [145.0, 149.0], "gt_se_norm": [0.6369383335113525, 0.6820952892303467], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005107.pt"}
{"idx": 5109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOaoI6-YAw4.mp4", "duration": 147.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a character mining with a pickaxe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a character mining with a pickaxe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 21.0], "base_se_norm": [-0.7551575899124146, -0.7143504619598389], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.7551575899124146, -0.7279528379440308], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005109.pt"}
{"idx": 5111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOaoI6-YAw4.mp4", "duration": 147.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the 'coffin dance' meme is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the 'coffin dance' meme is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 88.0], "base_se_norm": [0.07458615303039551, 0.197007417678833], "gt_se_raw": [80.0, 88.0], "gt_se_norm": [0.0881885290145874, 0.197007417678833], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005111.pt"}
{"idx": 5113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOaoI6-YAw4.mp4", "duration": 147.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the words 'this took way too long to record' appear on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the words 'this took way too long to record' appear on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.8911811113357544, -0.8639764189720154], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9047834873199463, -0.8775787949562073], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005113.pt"}
{"idx": 5115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r35hkbGnHgM.mp4", "duration": 26.292933333333334, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a sunrise is visible behind a row of windmills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a sunrise is visible behind a row of windmills.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.6957356929779053, -0.467537522315979], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.7718018293380737, -0.467537522315979], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005115.pt"}
{"idx": 5117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/r35hkbGnHgM.mp4", "duration": 26.292933333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where three swans are flying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where three swans are flying.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [0.06492495536804199, 0.29312312602996826], "gt_se_raw": [14.0, 17.0], "gt_se_norm": [0.06492495536804199, 0.29312312602996826], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005117.pt"}
{"idx": 5119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HTLgn4bZu3w.mp4", "duration": 61.89516666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a photo of a boy in a black hat is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a photo of a boy in a black hat is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.7091857194900513, -0.5799348950386047], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7091857194900513, -0.5799348950386047], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005119.pt"}
{"idx": 5121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HTLgn4bZu3w.mp4", "duration": 61.89516666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when a boy in a blue shirt is sitting next to a cooking appliance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a boy in a blue shirt is sitting next to a cooking appliance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 33.0], "base_se_norm": [-0.06293171644210815, 0.06631910800933838], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.0952444076538086, 0.06631910800933838], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005121.pt"}
{"idx": 5123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zywWM3J3i8M.mp4", "duration": 149.6495, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man walking with a white cane in a movie theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man walking with a white cane in a movie theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 58.0], "base_se_norm": [-0.27831369638442993, -0.2248554229736328], "gt_se_raw": [55.0, 58.0], "gt_se_norm": [-0.26494914293289185, -0.2248554229736328], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005123.pt"}
{"idx": 5125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zywWM3J3i8M.mp4", "duration": 149.6495, "prompt_text": "Please find the visual event described by the sentence 'At what point are the sounds of a lightsaber battle featured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the sounds of a lightsaber battle featured?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 85.0], "base_se_norm": [0.02907121181488037, 0.1359877586364746], "gt_se_raw": [77.0, 85.0], "gt_se_norm": [0.02907121181488037, 0.1359877586364746], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005125.pt"}
{"idx": 5127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zywWM3J3i8M.mp4", "duration": 149.6495, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where an example of the Descriptive Video Service is played.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where an example of the Descriptive Video Service is played.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 134.0], "base_se_norm": [0.6972992420196533, 0.7908512353897095], "gt_se_raw": [123.0, 128.0], "gt_se_norm": [0.643841028213501, 0.7106637954711914], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005127.pt"}
{"idx": 5129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v0jX8zWZaok.mp4", "duration": 309.684375, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man writes the word 'PIZZA' in the notebook.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man writes the word 'PIZZA' in the notebook.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 73.0], "base_se_norm": [-0.5608432292938232, -0.5285522937774658], "gt_se_raw": [69.0, 73.0], "gt_se_norm": [-0.5543850660324097, -0.5285522937774658], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005129.pt"}
{"idx": 5131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/v0jX8zWZaok.mp4", "duration": 309.684375, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man is playing the saxophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 132.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man is playing the saxophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 132.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 132.0], "base_se_norm": [-0.1733519434928894, -0.14751917123794556], "gt_se_raw": [129.0, 132.0], "gt_se_norm": [-0.16689372062683105, -0.14751917123794556], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005131.pt"}
{"idx": 5133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C7uAB94aRrQ.mp4", "duration": 29.988291666666665, "prompt_text": "Please find the visual event described by the sentence 'When does the puppy in the back seat bark at the truck driver?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppy in the back seat bark at the truck driver?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.3997657299041748, -0.199687659740448], "gt_se_raw": [10.0, 11.0], "gt_se_norm": [-0.3330730199813843, -0.26638031005859375], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005133.pt"}
{"idx": 5135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C7uAB94aRrQ.mp4", "duration": 29.988291666666665, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the puppy signals the truck driver to honk the horn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the puppy signals the truck driver to honk the horn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 13.0], "base_se_norm": [-0.26638031005859375, -0.13299494981765747], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.199687659740448, -0.13299494981765747], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005135.pt"}
{"idx": 5137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/C7uAB94aRrQ.mp4", "duration": 29.988291666666665, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the text 'The Subaru Forester' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the text 'The Subaru Forester' appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [0.4672393798828125, 0.7340101003646851], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.4672393798828125, 0.600624680519104], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005137.pt"}
{"idx": 5139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wAnrYQGSHMw.mp4", "duration": 175.71720833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man and woman doing push-ups with their feet on chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man and woman doing push-ups with their feet on chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 82.0], "base_se_norm": [-0.2260291576385498, -0.06668221950531006], "gt_se_raw": [68.0, 81.0], "gt_se_norm": [-0.2260291576385498, -0.07806414365768433], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005139.pt"}
{"idx": 5141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wAnrYQGSHMw.mp4", "duration": 175.71720833333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a woman is choosing a drink from the refrigerator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 153.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a woman is choosing a drink from the refrigerator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 153.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [153.0, 155.0], "base_se_norm": [0.7414344549179077, 0.7641983032226562], "gt_se_raw": [153.0, 155.0], "gt_se_norm": [0.7414344549179077, 0.7641983032226562], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005141.pt"}
{"idx": 5143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AoKlbyqbEGM.mp4", "duration": 32.76606666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman assisting the girl in playing the ukulele?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman assisting the girl in playing the ukulele?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 32.0], "base_se_norm": [-1.0, 0.953240156173706], "gt_se_raw": [0.0, 32.0], "gt_se_norm": [-1.0, 0.953240156173706], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005143.pt"}
{"idx": 5145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AoKlbyqbEGM.mp4", "duration": 32.76606666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl in the pink shirt raise her left hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl in the pink shirt raise her left hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 12.0], "base_se_norm": [-0.57272869348526, -0.26753491163253784], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [0.5870076417922974, 0.7701239585876465], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005145.pt"}
{"idx": 5147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AoKlbyqbEGM.mp4", "duration": 32.76606666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the girl looks up and smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the girl looks up and smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [0.6480463743209839, 0.953240156173706], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [0.7701239585876465, 0.953240156173706], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005147.pt"}
{"idx": 5149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1MsvTOTtr8A.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'Show me when the mascot is demonstrating push-ups in the gym.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the mascot is demonstrating push-ups in the gym.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [-0.26739931106567383, -0.06759911775588989], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [-0.26739931106567383, -0.06759911775588989], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005149.pt"}
{"idx": 5151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1MsvTOTtr8A.mp4", "duration": 60.059999999999995, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the mascot is washing its hands in front of a mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the mascot is washing its hands in front of a mirror.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 44.0], "base_se_norm": [0.2654012441635132, 0.46520137786865234], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [0.2321012020111084, 0.4319014549255371], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005151.pt"}
{"idx": 5153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3mG0XW9RoNM.mp4", "duration": 95.01158333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see scissors being used to cut the doll's long braid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see scissors being used to cut the doll's long braid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.9157997369766235], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.93684983253479], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005153.pt"}
{"idx": 5155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3mG0XW9RoNM.mp4", "duration": 95.01158333333332, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a needle-like tool is used to implant new, brown hair into a bald doll's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a needle-like tool is used to implant new, brown hair into a bald doll's head.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.7894993424415588, -0.6842490434646606], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.7263491153717041, -0.6842490434646606], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005155.pt"}
{"idx": 5157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3mG0XW9RoNM.mp4", "duration": 95.01158333333332, "prompt_text": "Please find the visual event described by the sentence 'When is a hair clipper used to shave off the doll's red, curly hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a hair clipper used to shave off the doll's red, curly hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 54.0], "base_se_norm": [0.010403156280517578, 0.1367034912109375], "gt_se_raw": [49.0, 54.0], "gt_se_norm": [0.03145325183868408, 0.1367034912109375], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005157.pt"}
{"idx": 5159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsVV0t_cS7Q.mp4", "duration": 119.54166666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the red t-shirt finish the challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the red t-shirt finish the challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 96.0], "base_se_norm": [0.5224816799163818, 0.6061345338821411], "gt_se_raw": [94.0, 96.0], "gt_se_norm": [0.5726734399795532, 0.6061345338821411], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005159.pt"}
{"idx": 5161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsVV0t_cS7Q.mp4", "duration": 119.54166666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the 'BADLANDS CHUGS' logo is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the 'BADLANDS CHUGS' logo is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 11.0], "base_se_norm": [-0.8996165990829468, -0.8159637451171875], "gt_se_raw": [5.0, 11.0], "gt_se_norm": [-0.9163471460342407, -0.8159637451171875], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005161.pt"}
{"idx": 5163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pNIGmCzL42g.mp4", "duration": 161.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the child start crying and putting their hand in their mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child start crying and putting their hand in their mouth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 14.0], "base_se_norm": [-0.9254966974258423, -0.8261589407920837], "gt_se_raw": [4.0, 13.0], "gt_se_norm": [-0.9503311514854431, -0.8385761380195618], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005163.pt"}
{"idx": 5165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pNIGmCzL42g.mp4", "duration": 161.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child rest their head on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child rest their head on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 91.0], "base_se_norm": [-0.006622493267059326, 0.1299668550491333], "gt_se_raw": [80.0, 88.0], "gt_se_norm": [-0.006622493267059326, 0.09271526336669922], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005165.pt"}
{"idx": 5167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pNIGmCzL42g.mp4", "duration": 161.06666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the child aggressively push the yellow box away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child aggressively push the yellow box away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.6274834275245667, -0.5902317762374878], "gt_se_raw": [143.0, 145.0], "gt_se_norm": [0.7756623029708862, 0.8004966974258423], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005167.pt"}
{"idx": 5169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ioq5yWKwE5o.mp4", "duration": 369.6, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the girl correctly points to the 'present' card on the black board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 194.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the girl correctly points to the 'present' card on the black board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 194.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [194.0, 198.0], "base_se_norm": [0.04978358745574951, 0.07142853736877441], "gt_se_raw": [137.0, 139.0], "gt_se_norm": [-0.2586579918861389, -0.24783551692962646], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005169.pt"}
{"idx": 5171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ioq5yWKwE5o.mp4", "duration": 369.6, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman reviews the three cards after placing them all on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman reviews the three cards after placing them all on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 92.0], "base_se_norm": [-0.7023809552192688, -0.5021644830703735], "gt_se_raw": [51.0, 55.0], "gt_se_norm": [-0.7240259647369385, -0.7023809552192688], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005171.pt"}
{"idx": 5173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RRt2YsawMj8.mp4", "duration": 99.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the teacher guiding the student's hands across the letter board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the teacher guiding the student's hands across the letter board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 13.0], "base_se_norm": [-1.0, -0.7386058568954468], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.8793565630912781, -0.7587131261825562], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005173.pt"}
{"idx": 5175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RRt2YsawMj8.mp4", "duration": 99.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the student rest her head on her arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the student rest her head on her arm?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 25.0], "base_se_norm": [-0.5978552103042603, -0.4973190426826477], "gt_se_raw": [32.0, 37.0], "gt_se_norm": [-0.3565683960914612, -0.25603216886520386], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005175.pt"}
{"idx": 5177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RRt2YsawMj8.mp4", "duration": 99.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the student spelling the word 'lot' by picking up the letter tiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the student spelling the word 'lot' by picking up the letter tiles?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 95.0], "base_se_norm": [0.34718501567840576, 0.9101876020431519], "gt_se_raw": [68.0, 94.0], "gt_se_norm": [0.3672921657562256, 0.890080451965332], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005177.pt"}
{"idx": 5179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10fmOuVnbco.mp4", "duration": 115.44866666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the time range when the blindfolded woman is assembling the red and yellow sphere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the time range when the blindfolded woman is assembling the red and yellow sphere.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 56.0], "base_se_norm": [-0.7574679851531982, -0.029871881008148193], "gt_se_raw": [20.0, 57.0], "gt_se_norm": [-0.653525710105896, -0.01254814863204956], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005179.pt"}
{"idx": 5181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10fmOuVnbco.mp4", "duration": 115.44866666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the black sweatshirt using tactile sign language on the other woman's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the black sweatshirt using tactile sign language on the other woman's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 112.0], "base_se_norm": [0.576458215713501, 0.9402562379837036], "gt_se_raw": [90.0, 112.0], "gt_se_norm": [0.5591344833374023, 0.9402562379837036], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005181.pt"}
{"idx": 5183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OUk4LxF5nzw.mp4", "duration": 405.0, "prompt_text": "Please find the visual event described by the sentence 'When does the car drift around a roundabout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the car drift around a roundabout?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 188.0], "base_se_norm": [-0.14567899703979492, -0.07160496711730957], "gt_se_raw": [154.0, 157.0], "gt_se_norm": [-0.23950618505477905, -0.22469133138656616], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005183.pt"}
{"idx": 5185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OUk4LxF5nzw.mp4", "duration": 405.0, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the car is driving inside a tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 323.0 - 333.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the car is driving inside a tunnel.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 323.0 - 333.0 seconds.<|im_end|>\n", "base_se_raw": [323.0, 333.0], "base_se_norm": [0.5950617790222168, 0.644444465637207], "gt_se_raw": [323.0, 329.0], "gt_se_norm": [0.5950617790222168, 0.624691367149353], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005185.pt"}
{"idx": 5187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OUk4LxF5nzw.mp4", "duration": 405.0, "prompt_text": "Please find the visual event described by the sentence 'When does the car accelerate rapidly, causing the tires to screech?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 318.0 - 323.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the car accelerate rapidly, causing the tires to screech?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 318.0 - 323.0 seconds.<|im_end|>\n", "base_se_raw": [318.0, 323.0], "base_se_norm": [0.5703703165054321, 0.5950617790222168], "gt_se_raw": [104.0, 106.0], "gt_se_norm": [-0.4864197373390198, -0.4765431880950928], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005187.pt"}
{"idx": 5189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JCOqo88eW1E.mp4", "duration": 33.9339, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl attempt to eat from the spoon by herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl attempt to eat from the spoon by herself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 12.0], "base_se_norm": [-0.7053093910217285, -0.29274260997772217], "gt_se_raw": [9.0, 12.0], "gt_se_norm": [-0.469556987285614, -0.29274260997772217], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005189.pt"}
{"idx": 5191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JCOqo88eW1E.mp4", "duration": 33.9339, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in the purple shirt smiling at the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in the purple shirt smiling at the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 33.0], "base_se_norm": [0.5913290977478027, 0.9449577331542969], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [0.5323909521102905, 0.6502672433853149], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005191.pt"}
{"idx": 5193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k0jstVk5AiU.mp4", "duration": 171.08758333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'HOW MUCH DO I HAVE IN LOANS?' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'HOW MUCH DO I HAVE IN LOANS?' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.6493024230003357, -0.6142326593399048], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.6493024230003357, -0.6142326593399048], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005193.pt"}
{"idx": 5195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k0jstVk5AiU.mp4", "duration": 171.08758333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the 'Submit' button on the loan calculator webpage clicked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the 'Submit' button on the loan calculator webpage clicked?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 133.0], "base_se_norm": [0.5313793420791626, 0.5547592639923096], "gt_se_raw": [130.0, 132.0], "gt_se_norm": [0.5196894407272339, 0.5430692434310913], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005195.pt"}
{"idx": 5197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k0jstVk5AiU.mp4", "duration": 171.08758333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the animation where papers for 'GRAD PLUS' and 'PARENT PLUS' loans are visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the animation where papers for 'GRAD PLUS' and 'PARENT PLUS' loans are visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 96.0], "base_se_norm": [0.040402770042419434, 0.12223219871520996], "gt_se_raw": [89.0, 97.0], "gt_se_norm": [0.040402770042419434, 0.13392210006713867], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005197.pt"}
{"idx": 5199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vvDIVVVAx-k.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker state that the employment rate for non-disabled individuals is 'much higher'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker state that the employment rate for non-disabled individuals is 'much higher'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9030095934867859, -0.6120384335517883], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.7090288400650024, -0.4180576801300049], "start_token_span": [75, 78], "end_token_span": [80, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005199.pt"}
{"idx": 5201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vvDIVVVAx-k.mp4", "duration": 20.6206, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where the data for the year 1991 is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where the data for the year 1991 is visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9398077726364136], "gt_se_raw": [0.0, 20.0], "gt_se_norm": [-1.0, 0.9398077726364136], "start_token_span": [79, 82], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005201.pt"}
{"idx": 5203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GT018Qejkhg.mp4", "duration": 122.6225, "prompt_text": "Please find the visual event described by the sentence 'When is the woman in the floral cardigan introducing herself and her job title?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman in the floral cardigan introducing herself and her job title?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 19.0], "base_se_norm": [-1.0, -0.6901057958602905], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, -0.8042773604393005], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005203.pt"}
{"idx": 5205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GT018Qejkhg.mp4", "duration": 122.6225, "prompt_text": "Please find the visual event described by the sentence 'At what point are three women seen discussing in front of a wall displaying shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are three women seen discussing in front of a wall displaying shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 122.0], "base_se_norm": [-0.4617627263069153, 0.98984694480896], "gt_se_raw": [32.0, 42.0], "gt_se_norm": [-0.4780729413032532, -0.3149707317352295], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005205.pt"}
{"idx": 5207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GT018Qejkhg.mp4", "duration": 122.6225, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a fourth woman in a dark top and colorful scarf briefly enters the scene from the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a fourth woman in a dark top and colorful scarf briefly enters the scene from the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 32.0], "base_se_norm": [-0.6737955808639526, -0.4780729413032532], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [0.2395768165588379, 0.27219724655151367], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005207.pt"}
{"idx": 5209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IANwb_qT1gg.mp4", "duration": 160.64, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a squirrel gets into the cockpit of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a squirrel gets into the cockpit of the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 84.0], "base_se_norm": [-0.06623506546020508, 0.045816779136657715], "gt_se_raw": [76.0, 78.0], "gt_se_norm": [-0.053784847259521484, -0.028884470462799072], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005209.pt"}
{"idx": 5211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IANwb_qT1gg.mp4", "duration": 160.64, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see the first-person view from the airplane with the squirrel in the cockpit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see the first-person view from the airplane with the squirrel in the cockpit?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 138.0], "base_se_norm": [0.593625545501709, 0.7181274890899658], "gt_se_raw": [128.0, 134.0], "gt_se_norm": [0.593625545501709, 0.668326735496521], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005211.pt"}
{"idx": 5213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ucYIwvZYGp8.mp4", "duration": 165.24841666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where a family of three enters their hotel suite for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where a family of three enters their hotel suite for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 26.0], "base_se_norm": [-0.7337342500686646, -0.6853222846984863], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.7337342500686646, -0.6974252462387085], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005213.pt"}
{"idx": 5215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ucYIwvZYGp8.mp4", "duration": 165.24841666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do people in the restaurant start hiding under the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do people in the restaurant start hiding under the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 71.0], "base_se_norm": [-0.16489362716674805, -0.14068764448165894], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.176996648311615, -0.15279066562652588], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005215.pt"}
{"idx": 5217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ucYIwvZYGp8.mp4", "duration": 165.24841666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man being warned over the phone not to open the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 119.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man being warned over the phone not to open the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 119.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [119.0, 122.0], "base_se_norm": [0.44025588035583496, 0.476564884185791], "gt_se_raw": [113.0, 115.0], "gt_se_norm": [0.3676379919052124, 0.39184391498565674], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005217.pt"}
{"idx": 5219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/URdX9rFIbcc.mp4", "duration": 193.56837499999997, "prompt_text": "Please find the visual event described by the sentence 'Show me where the pizza dough is being laid out on the baking sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me where the pizza dough is being laid out on the baking sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.3593994975090027, -0.32840269804000854], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.34906721115112305, -0.32840269804000854], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005219.pt"}
{"idx": 5221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/URdX9rFIbcc.mp4", "duration": 193.56837499999997, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the white sauce is being seasoned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the white sauce is being seasoned.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 41.0], "base_se_norm": [-0.6796997785568237, -0.5763770937919617], "gt_se_raw": [31.0, 36.0], "gt_se_norm": [-0.6796997785568237, -0.6280384063720703], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005221.pt"}
{"idx": 5223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4v22j4oiTsg.mp4", "duration": 430.13013013013017, "prompt_text": "Please find the visual event described by the sentence 'When is the butter being cut?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the butter being cut?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 23.0], "base_se_norm": [-0.9302536845207214, -0.8930556178092957], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.9302536845207214, -0.8977053761482239], "start_token_span": [61, 65], "end_token_span": [67, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005223.pt"}
{"idx": 5225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4v22j4oiTsg.mp4", "duration": 430.13013013013017, "prompt_text": "Please find the visual event described by the sentence 'At what point are the eggs being cracked into a bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the eggs being cracked into a bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 137.0], "base_se_norm": [-0.40483129024505615, -0.36298346519470215], "gt_se_raw": [128.0, 136.0], "gt_se_norm": [-0.40483129024505615, -0.36763322353363037], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005225.pt"}
{"idx": 5227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4v22j4oiTsg.mp4", "duration": 430.13013013013017, "prompt_text": "Please find the visual event described by the sentence 'When are the finished muffins being placed on the cooling rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 322.0 - 339.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the finished muffins being placed on the cooling rack?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 322.0 - 339.0 seconds.<|im_end|>\n", "base_se_raw": [322.0, 339.0], "base_se_norm": [0.49722135066986084, 0.5762671232223511], "gt_se_raw": [332.0, 341.0], "gt_se_norm": [0.5437189340591431, 0.5855666399002075], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005227.pt"}
{"idx": 5229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-BIL6p1Te8.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the paper is folded in half diagonally for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the paper is folded in half diagonally for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 47.0], "base_se_norm": [-0.6542836427688599, -0.4397010803222656], "gt_se_raw": [29.0, 37.0], "gt_se_norm": [-0.6542836427688599, -0.5589135885238647], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005229.pt"}
{"idx": 5231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z-BIL6p1Te8.mp4", "duration": 167.76760000000002, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the recommended paper size is shown as text.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the recommended paper size is shown as text.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 23.0], "base_se_norm": [-0.7973387241363525, -0.7258111834526062], "gt_se_raw": [17.0, 23.0], "gt_se_norm": [-0.7973387241363525, -0.7258111834526062], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005231.pt"}
{"idx": 5233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/akYVGzeS7_A.mp4", "duration": 85.11913333333332, "prompt_text": "Please find the visual event described by the sentence 'When is the woman's eye being removed and replaced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman's eye being removed and replaced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.8825175762176514, -0.7650352120399475], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.8825175762176514, -0.812028169631958], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005233.pt"}
{"idx": 5235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/akYVGzeS7_A.mp4", "duration": 85.11913333333332, "prompt_text": "Please find the visual event described by the sentence 'When is the woman being scanned by a blue light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman being scanned by a blue light?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.060140788555145264, 0.010348677635192871], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.060140788555145264, -0.013147830963134766], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005235.pt"}
{"idx": 5237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/akYVGzeS7_A.mp4", "duration": 85.11913333333332, "prompt_text": "Please find the visual event described by the sentence 'At what time does the red figure melt into a puddle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the red figure melt into a puddle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 84.0], "base_se_norm": [0.8797184228897095, 0.9737043380737305], "gt_se_raw": [79.0, 83.0], "gt_se_norm": [0.8562219142913818, 0.9502078294754028], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005237.pt"}
{"idx": 5239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R4WDWpR4oRM.mp4", "duration": 198.86533333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the red dress first walk up to the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the red dress first walk up to the man?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 51.0], "base_se_norm": [-0.517261266708374, -0.4870900511741638], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.517261266708374, -0.4971471428871155], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005239.pt"}
{"idx": 5241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R4WDWpR4oRM.mp4", "duration": 198.86533333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man faints and falls to the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 140.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man faints and falls to the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 140.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 140.0], "base_se_norm": [0.38787388801574707, 0.4079880714416504], "gt_se_raw": [138.0, 140.0], "gt_se_norm": [0.38787388801574707, 0.4079880714416504], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005241.pt"}
{"idx": 5243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DKoJm4OaZNw.mp4", "duration": 84.04, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the materials needed for this craft?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the materials needed for this craft?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.9286054372787476, -0.8334126472473145], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.9286054372787476, -0.8334126472473145], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005243.pt"}
{"idx": 5245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DKoJm4OaZNw.mp4", "duration": 84.04, "prompt_text": "Please find the visual event described by the sentence 'At what point are the four folded circles numbered to explain the steps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the four folded circles numbered to explain the steps?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.5478343963623047, -0.4526416063308716], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.5478343963623047, -0.4526416063308716], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005245.pt"}
{"idx": 5247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DKoJm4OaZNw.mp4", "duration": 84.04, "prompt_text": "Please find the visual event described by the sentence 'When is a yellow flower decoration added to the envelope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a yellow flower decoration added to the envelope?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [0.49928605556488037, 0.5706806182861328], "gt_se_raw": [63.0, 67.0], "gt_se_norm": [0.49928605556488037, 0.5944788455963135], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005247.pt"}
{"idx": 5249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10UKOhzoOws.mp4", "duration": 145.56, "prompt_text": "Please find the visual event described by the sentence 'When are the statistics about people with disabilities displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the statistics about people with disabilities displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 45.0], "base_se_norm": [-0.5190986394882202, -0.38169825077056885], "gt_se_raw": [35.0, 45.0], "gt_se_norm": [-0.5190986394882202, -0.38169825077056885], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005249.pt"}
{"idx": 5251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/10UKOhzoOws.mp4", "duration": 145.56, "prompt_text": "Please find the visual event described by the sentence 'At what point is the Google search interface shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the Google search interface shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 87.0], "base_se_norm": [0.08546304702758789, 0.19538331031799316], "gt_se_raw": [79.0, 87.0], "gt_se_norm": [0.08546304702758789, 0.19538331031799316], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005251.pt"}
{"idx": 5253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qw6RD5S3e8o.mp4", "duration": 30.3, "prompt_text": "Please find the visual event described by the sentence 'When can we see a crayon drawing of a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a crayon drawing of a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.40594059228897095], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, -0.40594059228897095], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005253.pt"}
{"idx": 5255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qw6RD5S3e8o.mp4", "duration": 30.3, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the red puppet is drawing on the yellow wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the red puppet is drawing on the yellow wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [0.2541254758834839, 0.7161716222763062], "gt_se_raw": [16.0, 21.0], "gt_se_norm": [0.056105613708496094, 0.3861386775970459], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005255.pt"}
{"idx": 5257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Qw6RD5S3e8o.mp4", "duration": 30.3, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the full 'Elmo's World' title is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the full 'Elmo's World' title is shown on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 29.0], "base_se_norm": [0.7161716222763062, 0.914191484451294], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [0.7821782827377319, 0.914191484451294], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005257.pt"}
{"idx": 5259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bxBHjpuXvCY.mp4", "duration": 71.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a teacher and a student are clapping hands at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a teacher and a student are clapping hands at a desk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.3599258065223694, -0.30426716804504395], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.3877550959587097, -0.33209651708602905], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005259.pt"}
{"idx": 5261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bxBHjpuXvCY.mp4", "duration": 71.86666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is a teacher showing a boy pictures on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is a teacher showing a boy pictures on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.7217068672180176, -0.6382189393043518], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.7495361566543579, -0.6382189393043518], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005261.pt"}
{"idx": 5263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EFeXMvjKWzo.mp4", "duration": 88.2, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an animated man is climbing up colorful steps towards a flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an animated man is climbing up colorful steps towards a flag.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 39.0], "base_se_norm": [-0.29705214500427246, -0.11564624309539795], "gt_se_raw": [31.0, 39.0], "gt_se_norm": [-0.29705214500427246, -0.11564624309539795], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005263.pt"}
{"idx": 5265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EFeXMvjKWzo.mp4", "duration": 88.2, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when a diagram illustrating the components of a CDR is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when a diagram illustrating the components of a CDR is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 31.0], "base_se_norm": [-0.45578229427337646, -0.29705214500427246], "gt_se_raw": [23.0, 31.0], "gt_se_norm": [-0.478458046913147, -0.29705214500427246], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005265.pt"}
{"idx": 5267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/EFeXMvjKWzo.mp4", "duration": 88.2, "prompt_text": "Please find the visual event described by the sentence 'At what time does the video explain the four-step order process?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the video explain the four-step order process?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 63.0], "base_se_norm": [0.2698413133621216, 0.4285714626312256], "gt_se_raw": [55.0, 63.0], "gt_se_norm": [0.24716556072235107, 0.4285714626312256], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005267.pt"}
{"idx": 5269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5jjeIH8Y6XM.mp4", "duration": 30.4304, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman is sweeping dirt into a dustpan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman is sweeping dirt into a dustpan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 9.0], "base_se_norm": [-0.5399338006973267, -0.4084862470626831], "gt_se_raw": [7.0, 8.0], "gt_se_norm": [-0.5399338006973267, -0.4742100238800049], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005269.pt"}
{"idx": 5271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5jjeIH8Y6XM.mp4", "duration": 30.4304, "prompt_text": "Please find the visual event described by the sentence 'Find the close-up shot of the dirty duster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the close-up shot of the dirty duster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.5116461515426636, 0.6430937051773071], "gt_se_raw": [20.0, 21.0], "gt_se_norm": [0.3144749402999878, 0.38019871711730957], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005271.pt"}
{"idx": 5273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KopKofh_-5E.mp4", "duration": 107.6075, "prompt_text": "Please find the visual event described by the sentence 'When are the old photos of women and cars shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the old photos of women and cars shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.7955532670021057, -0.7026229500770569], "gt_se_raw": [11.0, 15.0], "gt_se_norm": [-0.7955532670021057, -0.7212090492248535], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005273.pt"}
{"idx": 5275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KopKofh_-5E.mp4", "duration": 107.6075, "prompt_text": "Please find the visual event described by the sentence 'At what point does the camera focus on a black bunk bed frame installation in a gallery?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the camera focus on a black bunk bed frame installation in a gallery?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 27.0], "base_se_norm": [-0.5725204944610596, -0.4981762170791626], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.5725204944610596, -0.4981762170791626], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005275.pt"}
{"idx": 5277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KopKofh_-5E.mp4", "duration": 107.6075, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where three women are shown in separate black-and-white boxes, listening to music with earphones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where three women are shown in separate black-and-white boxes, listening to music with earphones.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 87.0], "base_se_norm": [0.5054713487625122, 0.6169877052307129], "gt_se_raw": [81.0, 87.0], "gt_se_norm": [0.5054713487625122, 0.6169877052307129], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005277.pt"}
{"idx": 5279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DfPMxdHZKsw.mp4", "duration": 99.91666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man consume the hot sauce?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man consume the hot sauce?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.01918262243270874, 0.06088411808013916], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [0.0008341073989868164, 0.06088411808013916], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005279.pt"}
{"idx": 5281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DfPMxdHZKsw.mp4", "duration": 99.91666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man first show the V8 Spicy Hot juice bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man first show the V8 Spicy Hot juice bottle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.9399499297142029, -0.8398665189743042], "gt_se_raw": [4.0, 7.0], "gt_se_norm": [-0.9199332594871521, -0.8598832488059998], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005281.pt"}
{"idx": 5283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkfLc-sQZNs.mp4", "duration": 94.9949, "prompt_text": "Please find the visual event described by the sentence 'When does the man stretch his arms above his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man stretch his arms above his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 30.0], "base_se_norm": [-0.49470973014831543, -0.3683871626853943], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.4736559987068176, -0.41049468517303467], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005283.pt"}
{"idx": 5285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkfLc-sQZNs.mp4", "duration": 94.9949, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man covers his face with his hand for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man covers his face with his hand for the second time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 59.0], "base_se_norm": [0.1369030475616455, 0.24217188358306885], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [0.1369030475616455, 0.22111809253692627], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005285.pt"}
{"idx": 5287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkfLc-sQZNs.mp4", "duration": 94.9949, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in the blue jacket imitate the man's stretching motion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in the blue jacket imitate the man's stretching motion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.4736559987068176, -0.3683871626853943], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.43154847621917725, -0.38944095373153687], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005287.pt"}
{"idx": 5289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aPH3XQbRdx4.mp4", "duration": 34.46776666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman say 'Good job' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman say 'Good job' for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 18.0], "base_se_norm": [-0.12962156534194946, 0.044454097747802734], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.5938234329223633, -0.47777295112609863], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005289.pt"}
{"idx": 5291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aPH3XQbRdx4.mp4", "duration": 34.46776666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the woman asks if he is going to knock on Simon's room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the woman asks if he is going to knock on Simon's room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 22.0], "base_se_norm": [0.044454097747802734, 0.27655506134033203], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.013571083545684814, 0.16050457954406738], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005291.pt"}
{"idx": 5293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99k-EAMBuM8.mp4", "duration": 51.44, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the duckling running from a concrete sidewalk to a wet asphalt road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the duckling running from a concrete sidewalk to a wet asphalt road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 16.0], "base_se_norm": [-0.5723172426223755, -0.37791597843170166], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.5723172426223755, -0.494556725025177], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005293.pt"}
{"idx": 5295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99k-EAMBuM8.mp4", "duration": 51.44, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person run past the duckling, causing it to turn and run back towards them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person run past the duckling, causing it to turn and run back towards them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 32.0], "base_se_norm": [0.010886549949645996, 0.24416804313659668], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [0.04976677894592285, 0.20528769493103027], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005295.pt"}
{"idx": 5297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/99k-EAMBuM8.mp4", "duration": 51.44, "prompt_text": "Please find the visual event described by the sentence 'When does the duckling run up to the camera and chirp loudly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the duckling run up to the camera and chirp loudly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 26.0], "base_se_norm": [-0.30015552043914795, 0.010886549949645996], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.30015552043914795, -0.1835147738456726], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005297.pt"}
{"idx": 5299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uOHvZjiDANg.mp4", "duration": 42.84, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cat tears up the drawings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cat tears up the drawings.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 32.0], "base_se_norm": [0.30718958377838135, 0.4939309358596802], "gt_se_raw": [28.0, 30.0], "gt_se_norm": [0.30718958377838135, 0.40056025981903076], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005299.pt"}
{"idx": 5301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uOHvZjiDANg.mp4", "duration": 42.84, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man get up from his desk to answer the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man get up from his desk to answer the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 26.0], "base_se_norm": [0.12044823169708252, 0.21381890773773193], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [0.12044823169708252, 0.21381890773773193], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005301.pt"}
{"idx": 5303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XVRLTfN1WYw.mp4", "duration": 89.1307, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone making the base of the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone making the base of the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 24.0], "base_se_norm": [-0.9102441668510437, -0.4614650011062622], "gt_se_raw": [4.0, 18.0], "gt_se_norm": [-0.9102441668510437, -0.596098780632019], "start_token_span": [72, 75], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005303.pt"}
{"idx": 5305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XVRLTfN1WYw.mp4", "duration": 89.1307, "prompt_text": "Please find the visual event described by the sentence 'When is the chocolate ice cream added to the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the chocolate ice cream added to the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.32683128118515015, -0.25951439142227173], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.32683128118515015, -0.25951439142227173], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005305.pt"}
{"idx": 5307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XVRLTfN1WYw.mp4", "duration": 89.1307, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone decorating the top of the cake slice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone decorating the top of the cake slice?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 65.0], "base_se_norm": [0.23414266109466553, 0.4585322141647339], "gt_se_raw": [56.0, 65.0], "gt_se_norm": [0.2565816640853882, 0.4585322141647339], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005307.pt"}
{"idx": 5309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/__n8LfdLDmU.mp4", "duration": 197.76423333333335, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man writes 'Pro 15:1' on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man writes 'Pro 15:1' on the board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [-0.23140805959701538, -0.1808428168296814], "gt_se_raw": [76.0, 81.0], "gt_se_norm": [-0.23140805959701538, -0.1808428168296814], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005309.pt"}
{"idx": 5311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/__n8LfdLDmU.mp4", "duration": 197.76423333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the man read from the book after writing on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man read from the book after writing on the board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 92.0], "base_se_norm": [-0.1403905749320984, -0.0695992112159729], "gt_se_raw": [84.0, 89.0], "gt_se_norm": [-0.15050363540649414, -0.09993839263916016], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005311.pt"}
{"idx": 5313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U_g910Y1ptw.mp4", "duration": 27.92, "prompt_text": "Please find the visual event described by the sentence 'When does the mother serve the soup to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the mother serve the soup to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.8567335605621338], "gt_se_raw": [1.0, 2.0], "gt_se_norm": [-0.9283667802810669, -0.8567335605621338], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005313.pt"}
{"idx": 5315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U_g910Y1ptw.mp4", "duration": 27.92, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the mother show her anger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the mother show her anger?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.2836676239967346, -0.06876790523529053], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [0.21776509284973145, 0.36103153228759766], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005315.pt"}
{"idx": 5317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U_g910Y1ptw.mp4", "duration": 27.92, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the girl is crying while eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the girl is crying while eating.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 27.0], "base_se_norm": [-0.06876790523529053, 0.934097409248352], "gt_se_raw": [22.0, 27.0], "gt_se_norm": [0.575931191444397, 0.934097409248352], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005317.pt"}
{"idx": 5319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DH2DgD9Rdi0.mp4", "duration": 60.166666666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the camera first pan to the dog and it sits there looking guilty?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera first pan to the dog and it sits there looking guilty?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 7.0], "base_se_norm": [-0.9335179924964905, -0.7673130035400391], "gt_se_raw": [2.0, 7.0], "gt_se_norm": [-0.9335179924964905, -0.7673130035400391], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005319.pt"}
{"idx": 5321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DH2DgD9Rdi0.mp4", "duration": 60.166666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the dog completely turn its head away to face the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the dog completely turn its head away to face the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.1689751148223877, -0.0027701258659362793], "gt_se_raw": [25.0, 33.0], "gt_se_norm": [-0.1689751148223877, 0.09695291519165039], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005321.pt"}
{"idx": 5323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghLkwSlWSXw.mp4", "duration": 197.03016666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the dad sneeze for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dad sneeze for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 17.0], "base_se_norm": [-0.8578897714614868, -0.8274375796318054], "gt_se_raw": [15.0, 17.0], "gt_se_norm": [-0.847739040851593, -0.8274375796318054], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005323.pt"}
{"idx": 5325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghLkwSlWSXw.mp4", "duration": 197.03016666666664, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy is running down the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy is running down the hallway.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.33005183935165405, -0.27929818630218506], "gt_se_raw": [65.0, 71.0], "gt_se_norm": [-0.34020256996154785, -0.27929818630218506], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005325.pt"}
{"idx": 5327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ghLkwSlWSXw.mp4", "duration": 197.03016666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the dad have a massive sneeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dad have a massive sneeze?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [-0.8375883102416992, -0.8172868490219116], "gt_se_raw": [174.0, 178.0], "gt_se_norm": [0.7662270069122314, 0.8068299293518066], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005327.pt"}
{"idx": 5329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SDeQT9zCvi4.mp4", "duration": 229.5293, "prompt_text": "Please find the visual event described by the sentence 'At what point is the donkey shown with only three legs and one eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 172.0 - 196.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the donkey shown with only three legs and one eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 172.0 - 196.0 seconds.<|im_end|>\n", "base_se_raw": [172.0, 196.0], "base_se_norm": [0.49871933460235596, 0.7078429460525513], "gt_se_raw": [36.0, 41.0], "gt_se_norm": [-0.686314582824707, -0.642747163772583], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005329.pt"}
{"idx": 5331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SDeQT9zCvi4.mp4", "duration": 229.5293, "prompt_text": "Please find the visual event described by the sentence 'When does the donkey break wind on the yellow bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the donkey break wind on the yellow bird?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 102.0], "base_se_norm": [-0.17221897840499878, -0.11122459173202515], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [-0.14607852697372437, -0.11122459173202515], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005331.pt"}
{"idx": 5333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vcgn4EY5_S8.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'When does the blonde woman use bolt cutters on the man's chain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blonde woman use bolt cutters on the man's chain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 62.0], "base_se_norm": [-0.4637620449066162, -0.36063939332962036], "gt_se_raw": [52.0, 62.0], "gt_se_norm": [-0.4637620449066162, -0.36063939332962036], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005333.pt"}
{"idx": 5335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vcgn4EY5_S8.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'At what point are all three main characters (Max, Furiosa, Nux) wrestling on the ground together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are all three main characters (Max, Furiosa, Nux) wrestling on the ground together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 98.0], "base_se_norm": [-0.25751668214797974, 0.010602235794067383], "gt_se_raw": [157.0, 164.0], "gt_se_norm": [0.6190260648727417, 0.6912119388580322], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005335.pt"}
{"idx": 5337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Vcgn4EY5_S8.mp4", "duration": 193.94375, "prompt_text": "Please find the visual event described by the sentence 'Show me when the women are pulling on the chain to drag the man across the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the women are pulling on the chain to drag the man across the sand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 106.0], "base_se_norm": [-0.36063939332962036, 0.09310042858123779], "gt_se_raw": [77.0, 80.0], "gt_se_norm": [-0.2059553861618042, -0.1750185489654541], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005337.pt"}
{"idx": 5339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0WWzgGyAH6Y.mp4", "duration": 219.17729166666666, "prompt_text": "Please find the visual event described by the sentence 'When does the rocket explode after launch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the rocket explode after launch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 102.0], "base_se_norm": [-0.1148717999458313, -0.06924664974212646], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [-0.1148717999458313, -0.08749669790267944], "start_token_span": [62, 66], "end_token_span": [68, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005339.pt"}
{"idx": 5341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/0WWzgGyAH6Y.mp4", "duration": 219.17729166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does Batman turn on the Bat-Signal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Batman turn on the Bat-Signal?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 136.0], "base_se_norm": [0.2045043706893921, 0.24100446701049805], "gt_se_raw": [132.0, 136.0], "gt_se_norm": [0.2045043706893921, 0.24100446701049805], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005341.pt"}
{"idx": 5343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yq3Z9msqnOg.mp4", "duration": 162.625, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the bruise on the boy's hand transforms into an angry face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the bruise on the boy's hand transforms into an angry face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [-0.31129902601242065, -0.27440428733825684], "gt_se_raw": [63.0, 66.0], "gt_se_norm": [-0.22521138191223145, -0.1883167028427124], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005343.pt"}
{"idx": 5345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yq3Z9msqnOg.mp4", "duration": 162.625, "prompt_text": "Please find the visual event described by the sentence 'When does the girl start hugging the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl start hugging the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 134.0], "base_se_norm": [0.5987701416015625, 0.6479630470275879], "gt_se_raw": [129.0, 133.0], "gt_se_norm": [0.5864719152450562, 0.6356648206710815], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005345.pt"}
{"idx": 5347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yq3Z9msqnOg.mp4", "duration": 162.625, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the teacher is standing in front of the whiteboard, which has a math problem on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the teacher is standing in front of the whiteboard, which has a math problem on it.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [-0.0530361533164978, -0.0038431882858276367], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [-0.04073786735534668, -0.0038431882858276367], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005347.pt"}
{"idx": 5349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ER8yEOY0NCc.mp4", "duration": 49.3493, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the woman wipes the child's chin with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the woman wipes the child's chin with her hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.8378903269767761, -0.7568354606628418], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8784177303314209, -0.7973628640174866], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005349.pt"}
{"idx": 5351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ER8yEOY0NCc.mp4", "duration": 49.3493, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman feeds the child a spoonful of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman feeds the child a spoonful of food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 6.0], "base_se_norm": [-0.9189451336860657, -0.7568354606628418], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [0.41845977306365967, 0.49951469898223877], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005351.pt"}
{"idx": 5353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/foN-gLb27jY.mp4", "duration": 153.754, "prompt_text": "Please find the visual event described by the sentence 'When does the man say 'cookie time'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say 'cookie time'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.6748051643371582, -0.6357818245887756], "gt_se_raw": [15.0, 19.0], "gt_se_norm": [-0.8048831224441528, -0.7528519630432129], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005353.pt"}
{"idx": 5355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/foN-gLb27jY.mp4", "duration": 153.754, "prompt_text": "Please find the visual event described by the sentence 'Show me when the man and the boy are playing drums on the baking sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the man and the boy are playing drums on the baking sheet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 84.0], "base_se_norm": [-0.05043119192123413, 0.09265446662902832], "gt_se_raw": [62.0, 83.0], "gt_se_norm": [-0.19351691007614136, 0.07964670658111572], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005355.pt"}
{"idx": 5357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/foN-gLb27jY.mp4", "duration": 153.754, "prompt_text": "Please find the visual event described by the sentence 'When does the man put the sugar into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man put the sugar into the bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.6748051643371582, -0.6097662448883057], "gt_se_raw": [129.0, 141.0], "gt_se_norm": [0.6780050992965698, 0.8340986967086792], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005357.pt"}
{"idx": 5359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wEp4ItQzzCU.mp4", "duration": 129.26246666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the puppy is lying on the floor chewing on a person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the puppy is lying on the floor chewing on a person's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 49.0], "base_se_norm": [-0.3037422299385071, -0.24185264110565186], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [-0.3037422299385071, -0.2573249936103821], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005359.pt"}
{"idx": 5361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wEp4ItQzzCU.mp4", "duration": 129.26246666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the puppy walk away towards the left side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the puppy walk away towards the left side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.6596072912216187, -0.5822453498840332], "gt_se_raw": [99.0, 102.0], "gt_se_norm": [0.5317671298980713, 0.5781843662261963], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005361.pt"}
{"idx": 5363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZgwNixjNyzI.mp4", "duration": 29.829, "prompt_text": "Please find the visual event described by the sentence 'At what time is the girl with blonde hair reading a book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the girl with blonde hair reading a book?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.7988534569740295], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.865902304649353], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005363.pt"}
{"idx": 5365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZgwNixjNyzI.mp4", "duration": 29.829, "prompt_text": "Please find the visual event described by the sentence 'When can we see a boy brushing his teeth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a boy brushing his teeth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 19.0], "base_se_norm": [-0.06131619215011597, 0.27392804622650146], "gt_se_raw": [15.0, 18.0], "gt_se_norm": [0.0057326555252075195, 0.2068791389465332], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005365.pt"}
{"idx": 5367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZgwNixjNyzI.mp4", "duration": 29.829, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where a girl is running with a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where a girl is running with a kite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [0.5421234369277954, 0.6762211322784424], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.5421234369277954, 0.6762211322784424], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005367.pt"}
{"idx": 5369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUeiWQtcPYM.mp4", "duration": 113.48014681348015, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a woman is working on a computer, displaying a 3D interior design rendering.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a woman is working on a computer, displaying a 3D interior design rendering.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.5593942999839783, -0.5065215826034546], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.5593942999839783, -0.5065215826034546], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005369.pt"}
{"idx": 5371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cUeiWQtcPYM.mp4", "duration": 113.48014681348015, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a waiter is serving food to customers at an outdoor table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a waiter is serving food to customers at an outdoor table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 56.0], "base_se_norm": [-0.048291683197021484, -0.013043224811553955], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [-0.06591594219207764, -0.013043224811553955], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005371.pt"}
{"idx": 5373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F5B9jmB_4oM.mp4", "duration": 136.46966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see two men riding horses and a huge explosion in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see two men riding horses and a huge explosion in the background?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 100.0], "base_se_norm": [0.4362165927886963, 0.46552717685699463], "gt_se_raw": [99.0, 101.0], "gt_se_norm": [0.4508718252182007, 0.480182409286499], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005373.pt"}
{"idx": 5375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F5B9jmB_4oM.mp4", "duration": 136.46966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man wearing a hood and mask take aim with a bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man wearing a hood and mask take aim with a bow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 80.0], "base_se_norm": [0.09914541244506836, 0.17242169380187988], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.48706549406051636, -0.4577549695968628], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005375.pt"}
{"idx": 5377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/F5B9jmB_4oM.mp4", "duration": 136.46966666666665, "prompt_text": "Please find the visual event described by the sentence 'When can we see the logo of a movie production company called 'SUMMIT ENTERTAINMENT'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the logo of a movie production company called 'SUMMIT ENTERTAINMENT'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [-0.7655156850814819, -0.7362051010131836], "gt_se_raw": [16.0, 18.0], "gt_se_norm": [-0.7655156850814819, -0.7362051010131836], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005377.pt"}
{"idx": 5379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CU7VAIup27U.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where a girl is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where a girl is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 16.0], "base_se_norm": [-0.5333333015441895, -0.46666663885116577], "gt_se_raw": [14.0, 15.0], "gt_se_norm": [-0.5333333015441895, -0.5], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005379.pt"}
{"idx": 5381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CU7VAIup27U.mp4", "duration": 60.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment someone is serving the roast dinner to the family at the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment someone is serving the roast dinner to the family at the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [0.3333333730697632, 0.6000000238418579], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [0.36666667461395264, 0.46666669845581055], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005381.pt"}
{"idx": 5383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zd2xn5U6e-E.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'Show me when the nanny in the blue coat and red hat enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the nanny in the blue coat and red hat enters the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 24.0], "base_se_norm": [-0.8199282884597778, -0.7457810640335083], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.8093358278274536, -0.7881509065628052], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005383.pt"}
{"idx": 5385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zd2xn5U6e-E.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the two men are pretending to have rotten teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the two men are pretending to have rotten teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 82.0], "base_se_norm": [-0.15260356664657593, -0.1314186453819275], "gt_se_raw": [80.0, 82.0], "gt_se_norm": [-0.15260356664657593, -0.1314186453819275], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005385.pt"}
{"idx": 5387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zd2xn5U6e-E.mp4", "duration": 188.81362499999997, "prompt_text": "Please find the visual event described by the sentence 'When does the nanny point her finger and shout at the people on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the nanny point her finger and shout at the people on the couch?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 94.0], "base_se_norm": [-0.04667896032333374, -0.004309177398681641], "gt_se_raw": [166.0, 168.0], "gt_se_norm": [0.7583476305007935, 0.7795325517654419], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005387.pt"}
{"idx": 5389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eaIvk1cSyG8.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy start giving advice to other kids?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy start giving advice to other kids?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 47.0], "base_se_norm": [-0.3337445855140686, 0.7396669387817383], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.2227020263671875, -0.1116594672203064], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005389.pt"}
{"idx": 5391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/eaIvk1cSyG8.mp4", "duration": 54.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy gives two thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy gives two thumbs up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.7766810655593872, 0.924737811088562], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [0.7396669387817383, 0.8136953115463257], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005391.pt"}
{"idx": 5393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/17W4QMNwm-c.mp4", "duration": 89.55613333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the two men engaged in a lightsaber duel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men engaged in a lightsaber duel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 48.0], "base_se_norm": [-0.7543440461158752, 0.07195329666137695], "gt_se_raw": [11.0, 48.0], "gt_se_norm": [-0.7543440461158752, 0.07195329666137695], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005393.pt"}
{"idx": 5395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/17W4QMNwm-c.mp4", "duration": 89.55613333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man with the red lightsaber get on his white horse and ride away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man with the red lightsaber get on his white horse and ride away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 76.0], "base_se_norm": [0.5855976343154907, 0.6972594261169434], "gt_se_raw": [71.0, 77.0], "gt_se_norm": [0.5855976343154907, 0.7195918560028076], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005395.pt"}
{"idx": 5397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/17W4QMNwm-c.mp4", "duration": 89.55613333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the man in the light-colored jacket striking a soldier in armor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man in the light-colored jacket striking a soldier in armor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.7543440461158752, -0.6650146245956421], "gt_se_raw": [57.0, 59.0], "gt_se_norm": [0.2729445695877075, 0.3176093101501465], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005397.pt"}
{"idx": 5399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z3ZAGBL6UBA.mp4", "duration": 106.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time does the banana crouch down and dance low to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the banana crouch down and dance low to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 62.0], "base_se_norm": [-0.07619798183441162, 0.16889238357543945], "gt_se_raw": [44.0, 50.0], "gt_se_norm": [-0.17046350240707397, -0.05734485387802124], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005399.pt"}
{"idx": 5401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Z3ZAGBL6UBA.mp4", "duration": 106.08333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the lyrics say 'Tic-tac-toe'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the lyrics say 'Tic-tac-toe'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.43440693616867065, -0.3778476119041443], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [0.3008641004562378, 0.39512956142425537], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005401.pt"}
{"idx": 5403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CrqIVVd3hp0.mp4", "duration": 154.6, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the wheelchair receive an invitation on his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the wheelchair receive an invitation on his phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.9741267561912537, -0.9353169202804565], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9870634078979492, -0.9482535719871521], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005403.pt"}
{"idx": 5405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CrqIVVd3hp0.mp4", "duration": 154.6, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman read about the 'LECTURE BY FAMOUS SPEAKER TONIGHT' in the newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman read about the 'LECTURE BY FAMOUS SPEAKER TONIGHT' in the newspaper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.9223803281784058, -0.8706339001655579], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.909443736076355, -0.8835704922676086], "start_token_span": [80, 83], "end_token_span": [85, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005405.pt"}
{"idx": 5407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CrqIVVd3hp0.mp4", "duration": 154.6, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the four main characters standing together at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the four main characters standing together at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 151.0], "base_se_norm": [0.901681661605835, 0.9534281492233276], "gt_se_raw": [148.0, 151.0], "gt_se_norm": [0.9146182537078857, 0.9534281492233276], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005407.pt"}
{"idx": 5409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MId2A3Mdyn0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man in a blue polo shirt write on a glowing board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man in a blue polo shirt write on a glowing board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5409404635429382, -0.08188092708587646], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005409.pt"}
{"idx": 5411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MId2A3Mdyn0.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman in a purple t-shirt turns and smiles at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman in a purple t-shirt turns and smiles at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005411.pt"}
{"idx": 5413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i3kIpCzLzEo.mp4", "duration": 54.053999999999995, "prompt_text": "Please find the visual event described by the sentence 'When does the baby fish say she has short-term memory loss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the baby fish say she has short-term memory loss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 36.0], "base_se_norm": [0.22100114822387695, 0.33200132846832275], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9629999399185181, -0.8519998788833618], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005413.pt"}
{"idx": 5415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i3kIpCzLzEo.mp4", "duration": 54.053999999999995, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the two adult fish hide behind the pink seaweed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the two adult fish hide behind the pink seaweed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 34.0], "base_se_norm": [-0.11199915409088135, 0.2580012083053589], "gt_se_raw": [23.0, 27.0], "gt_se_norm": [-0.1489991545677185, -0.0009990334510803223], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005415.pt"}
{"idx": 5417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i3kIpCzLzEo.mp4", "duration": 54.053999999999995, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the baby fish swims off and her parents call her name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the baby fish swims off and her parents call her name.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 48.0], "base_se_norm": [0.18400120735168457, 0.7760016918182373], "gt_se_raw": [45.0, 49.0], "gt_se_norm": [0.665001630783081, 0.8130017518997192], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005417.pt"}
{"idx": 5419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uEu6r8MkQ0o.mp4", "duration": 152.08541875208542, "prompt_text": "Please find the visual event described by the sentence 'At what time is the red tent with 'NEW BELGIUM' on it visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the red tent with 'NEW BELGIUM' on it visible?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.6712373495101929, -0.6317858695983887], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.658086895942688, -0.6317858695983887], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005419.pt"}
{"idx": 5421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uEu6r8MkQ0o.mp4", "duration": 152.08541875208542, "prompt_text": "Please find the visual event described by the sentence 'Show me the timestamp for a drink being poured from a tap into a wine glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the timestamp for a drink being poured from a tap into a wine glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.13206666707992554, -0.06631416082382202], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [-0.13206666707992554, -0.07946467399597168], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005421.pt"}
{"idx": 5423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tDRBb0apDg.mp4", "duration": 154.02053333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black shirt point a gun at his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black shirt point a gun at his head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.83119136095047, -0.7922354936599731], "gt_se_raw": [12.0, 16.0], "gt_se_norm": [-0.8441766500473022, -0.7922354936599731], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005423.pt"}
{"idx": 5425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tDRBb0apDg.mp4", "duration": 154.02053333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the man in the grey shirt holds up a picture of Sirius Black.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the man in the grey shirt holds up a picture of Sirius Black.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 63.0], "base_se_norm": [-0.272824227809906, -0.18192726373672485], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [-0.2598389983177185, -0.18192726373672485], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005425.pt"}
{"idx": 5427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tDRBb0apDg.mp4", "duration": 154.02053333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do the two men run across the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the two men run across the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 99.0], "base_se_norm": [0.23360168933868408, 0.2855428457260132], "gt_se_raw": [96.0, 99.0], "gt_se_norm": [0.24658703804016113, 0.2855428457260132], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005427.pt"}
{"idx": 5429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VFbYadm_mrw.mp4", "duration": 122.03858333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is walking and playing with his puppy in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is walking and playing with his puppy in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.606681764125824, -0.47557568550109863], "gt_se_raw": [25.0, 31.0], "gt_se_norm": [-0.5902935266494751, -0.4919639229774475], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005429.pt"}
{"idx": 5431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VFbYadm_mrw.mp4", "duration": 122.03858333333332, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the 'Campfire' search results page is displayed, including the 'Smell' button.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the 'Campfire' search results page is displayed, including the 'Smell' button.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 65.0], "base_se_norm": [-0.04948091506958008, 0.06523692607879639], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.0330926775932312, 0.06523692607879639], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005431.pt"}
{"idx": 5433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sNcJMejrcnM.mp4", "duration": 104.23746666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker suggest that the tray can function as a social obstacle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker suggest that the tray can function as a social obstacle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 60.0], "base_se_norm": [-0.2517085671424866, 0.15121757984161377], "gt_se_raw": [40.0, 60.0], "gt_se_norm": [-0.232521653175354, 0.15121757984161377], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005433.pt"}
{"idx": 5435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sNcJMejrcnM.mp4", "duration": 104.23746666666666, "prompt_text": "Please find the visual event described by the sentence 'When is an example of items being placed on the boy's tray shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is an example of items being placed on the boy's tray shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 39.0], "base_se_norm": [-0.4627651572227478, -0.2517085671424866], "gt_se_raw": [28.0, 40.0], "gt_se_norm": [-0.4627651572227478, -0.232521653175354], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005435.pt"}
{"idx": 5437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sNcJMejrcnM.mp4", "duration": 104.23746666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the boy is using a table just like his peers, without the wheelchair tray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the boy is using a table just like his peers, without the wheelchair tray.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 104.0], "base_se_norm": [0.4390219449996948, 0.9954438209533691], "gt_se_raw": [74.0, 104.0], "gt_se_norm": [0.4198349714279175, 0.9954438209533691], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005437.pt"}
{"idx": 5439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4GfBVEoxStA.mp4", "duration": 156.48966666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the slide that discusses the movie 'The Hunchback of Notre Dame'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the slide that discusses the movie 'The Hunchback of Notre Dame'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 92.0], "base_se_norm": [-0.41210174560546875, 0.1757965087890625], "gt_se_raw": [45.0, 92.0], "gt_se_norm": [-0.424882173538208, 0.1757965087890625], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005439.pt"}
{"idx": 5441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4GfBVEoxStA.mp4", "duration": 156.48966666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker mention that Nemo has an undersized fin but calls it 'lucky'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker mention that Nemo has an undersized fin but calls it 'lucky'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 102.0], "base_se_norm": [0.1885768175125122, 0.3036004304885864], "gt_se_raw": [95.0, 100.0], "gt_se_norm": [0.21413767337799072, 0.2780395746231079], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005441.pt"}
{"idx": 5443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MdEV-jWMGWw.mp4", "duration": 190.52385719052387, "prompt_text": "Please find the visual event described by the sentence 'When are the man and woman in the plane on the runway before it takes off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the man and woman in the plane on the runway before it takes off?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 25.0], "base_se_norm": [-0.9580104947090149, -0.7375656962394714], "gt_se_raw": [4.0, 17.0], "gt_se_norm": [-0.9580104947090149, -0.8215446472167969], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005443.pt"}
{"idx": 5445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MdEV-jWMGWw.mp4", "duration": 190.52385719052387, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man take out the ring and propose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man take out the ring and propose?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 154.0], "base_se_norm": [0.37515580654144287, 0.6165953874588013], "gt_se_raw": [126.0, 153.0], "gt_se_norm": [0.3226689100265503, 0.6060980558395386], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005445.pt"}
{"idx": 5447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MdEV-jWMGWw.mp4", "duration": 190.52385719052387, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman is talking to the camera on the ground after landing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 190.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman is talking to the camera on the ground after landing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 190.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 190.0], "base_se_norm": [0.8580349683761597, 0.9945007562637329], "gt_se_raw": [177.0, 190.0], "gt_se_norm": [0.8580349683761597, 0.9945007562637329], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005447.pt"}
{"idx": 5449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pz70gsQVr40.mp4", "duration": 188.6050833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the Lego-shaped candies poured into the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 153.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the Lego-shaped candies poured into the drawer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 153.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [153.0, 156.0], "base_se_norm": [0.6224377155303955, 0.6542501449584961], "gt_se_raw": [153.0, 155.0], "gt_se_norm": [0.6224377155303955, 0.6436460018157959], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005449.pt"}
{"idx": 5451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pz70gsQVr40.mp4", "duration": 188.6050833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the woman holds up the Coke bottle-shaped gummies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the woman holds up the Coke bottle-shaped gummies.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [-0.18347907066345215, -0.13045823574066162], "gt_se_raw": [78.0, 81.0], "gt_se_norm": [-0.17287492752075195, -0.14106237888336182], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005451.pt"}
{"idx": 5453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtP83CWOMwc.mp4", "duration": 198.45833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the three little pigs, in different colored overalls, are introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the three little pigs, in different colored overalls, are introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 11.0], "base_se_norm": [-0.9697669744491577, -0.8891454935073853], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.9596892595291138, -0.8891454935073853], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005453.pt"}
{"idx": 5455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtP83CWOMwc.mp4", "duration": 198.45833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the pig in yellow overalls flee to his brother's house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the pig in yellow overalls flee to his brother's house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 90.0], "base_se_norm": [-0.13331931829452515, -0.09300857782363892], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [-0.13331931829452515, -0.10308629274368286], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005455.pt"}
{"idx": 5457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CtP83CWOMwc.mp4", "duration": 198.45833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the wolf falls into the boiling water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 166.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the wolf falls into the boiling water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 166.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 166.0], "base_se_norm": [0.6325845718383789, 0.6728953123092651], "gt_se_raw": [161.0, 165.0], "gt_se_norm": [0.622506856918335, 0.6628175973892212], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005457.pt"}
{"idx": 5459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLeZc7li5HM.mp4", "duration": 96.096, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher touch the student's shoulder to praise him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher touch the student's shoulder to praise him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.6045621037483215, -0.5629370808601379], "gt_se_raw": [19.0, 20.0], "gt_se_norm": [-0.6045621037483215, -0.5837495923042297], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005459.pt"}
{"idx": 5461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jLeZc7li5HM.mp4", "duration": 96.096, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the student places his hand near his ear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the student places his hand near his ear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 62.0], "base_se_norm": [0.12387609481811523, 0.29037630558013916], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [0.14468860626220703, 0.18631362915039062], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005461.pt"}
{"idx": 5463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZG4coPHkTmQ.mp4", "duration": 112.57913333333333, "prompt_text": "Please find the visual event described by the sentence 'When do three eggs appear in the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do three eggs appear in the nest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 30.0], "base_se_norm": [-0.6446943283081055, -0.467041552066803], "gt_se_raw": [20.0, 30.0], "gt_se_norm": [-0.6446943283081055, -0.467041552066803], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005463.pt"}
{"idx": 5465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZG4coPHkTmQ.mp4", "duration": 112.57913333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the bird blinks during the close-up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the bird blinks during the close-up.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.12950122356414795, -0.0939706563949585], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [-0.16503173112869263, -0.12950122356414795], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005465.pt"}
{"idx": 5467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZG4coPHkTmQ.mp4", "duration": 112.57913333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the nest with the bird in it fall off the cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the nest with the bird in it fall off the cliff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 92.0], "base_se_norm": [0.4389878511428833, 0.6344059705734253], "gt_se_raw": [89.0, 92.0], "gt_se_norm": [0.5811101198196411, 0.6344059705734253], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005467.pt"}
{"idx": 5469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/glQ2NrnDHWc.mp4", "duration": 180.16, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where a pink balloon is being pushed down onto multiple pins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where a pink balloon is being pushed down onto multiple pins.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 49.0], "base_se_norm": [-0.644760251045227, -0.45603907108306885], "gt_se_raw": [34.0, 39.0], "gt_se_norm": [-0.6225577592849731, -0.5670515298843384], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005469.pt"}
{"idx": 5471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/glQ2NrnDHWc.mp4", "duration": 180.16, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the green balloon self-inflates from the chemical reaction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the green balloon self-inflates from the chemical reaction.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 102.0], "base_se_norm": [-0.056394338607788086, 0.13232684135437012], "gt_se_raw": [88.0, 98.0], "gt_se_norm": [-0.023090600967407227, 0.0879218578338623], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005471.pt"}
{"idx": 5473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G4Sn91t1V4g.mp4", "duration": 177.09358333333333, "prompt_text": "Please find the visual event described by the sentence 'A cat is shown wearing a lion mane costume.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A cat is shown wearing a lion mane costume.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [-0.7515437602996826, -0.7289568185806274], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.762837290763855, -0.7402503490447998], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005473.pt"}
{"idx": 5475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G4Sn91t1V4g.mp4", "duration": 177.09358333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a person opens a can of cat food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a person opens a can of cat food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 102.0], "base_se_norm": [0.12934637069702148, 0.15193331241607666], "gt_se_raw": [100.0, 102.0], "gt_se_norm": [0.12934637069702148, 0.15193331241607666], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005475.pt"}
{"idx": 5477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G4Sn91t1V4g.mp4", "duration": 177.09358333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the little kitten chase a red laser dot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little kitten chase a red laser dot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 153.0], "base_se_norm": [0.5584980249404907, 0.7279000282287598], "gt_se_raw": [137.0, 153.0], "gt_se_norm": [0.5472046136856079, 0.7279000282287598], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005477.pt"}
{"idx": 5479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rUdA54Xk8cg.mp4", "duration": 106.02258333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a branch with leaves sprouts from Pinocchio's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a branch with leaves sprouts from Pinocchio's nose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 33.0], "base_se_norm": [-0.5095384120941162, -0.3774911165237427], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.5095384120941162, -0.47181063890457153], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005479.pt"}
{"idx": 5481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/rUdA54Xk8cg.mp4", "duration": 106.02258333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the leaves on the branch wither and fall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the leaves on the branch wither and fall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [0.16956210136413574, 0.24501776695251465], "gt_se_raw": [62.0, 64.0], "gt_se_norm": [0.16956210136413574, 0.2072899341583252], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005481.pt"}
{"idx": 5483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U9Ots7t3_4g.mp4", "duration": 199.63276666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the people singing about what day of the week it is?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the people singing about what day of the week it is?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 46.0], "base_se_norm": [-0.6593745946884155, -0.5391538143157959], "gt_se_raw": [2.0, 15.0], "gt_se_norm": [-0.9799631834030151, -0.8497240543365479], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005483.pt"}
{"idx": 5485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U9Ots7t3_4g.mp4", "duration": 199.63276666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the child is being redirected from scratching or biting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 105.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the child is being redirected from scratching or biting.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 105.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [105.0, 119.0], "base_se_norm": [0.05193150043487549, 0.19218909740447998], "gt_se_raw": [86.0, 96.0], "gt_se_norm": [-0.1384180188179016, -0.038234055042266846], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005485.pt"}
{"idx": 5487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U9Ots7t3_4g.mp4", "duration": 199.63276666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the group sings 'Hello and good morning' to Samantha.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the group sings 'Hello and good morning' to Samantha.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 188.0], "base_se_norm": [0.6730719804763794, 0.8834583759307861], "gt_se_raw": [147.0, 155.0], "gt_se_norm": [0.4727041721343994, 0.5528513193130493], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005487.pt"}
{"idx": 5489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z1fbwPHv-wA.mp4", "duration": 68.9689, "prompt_text": "Please find the visual event described by the sentence 'At what point does the boy drop the cane on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the boy drop the cane on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.3330327272415161, -0.24603700637817383], "gt_se_raw": [24.0, 25.0], "gt_se_norm": [-0.30403417348861694, -0.275035560131073], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005489.pt"}
{"idx": 5491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/z1fbwPHv-wA.mp4", "duration": 68.9689, "prompt_text": "Please find the visual event described by the sentence 'Identify the scene where the boy looks up to speak with the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the scene where the boy looks up to speak with the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [0.3919316530227661, 0.5079259872436523], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.3919316530227661, 0.5079259872436523], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005491.pt"}
{"idx": 5493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f-B-640jgWY.mp4", "duration": 166.53303333333335, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the man is sitting in front of a bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the man is sitting in front of a bookshelf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 28.0], "base_se_norm": [-0.9399518370628357, -0.6637303829193115], "gt_se_raw": [5.0, 29.0], "gt_se_norm": [-0.9399518370628357, -0.6517207622528076], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005493.pt"}
{"idx": 5495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f-B-640jgWY.mp4", "duration": 166.53303333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me the man in a tie-dye shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the man in a tie-dye shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 136.0], "base_se_norm": [0.5972806215286255, 0.6333096027374268], "gt_se_raw": [133.0, 136.0], "gt_se_norm": [0.5972806215286255, 0.6333096027374268], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005495.pt"}
{"idx": 5497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f-B-640jgWY.mp4", "duration": 166.53303333333335, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment all the speakers' video feeds are displayed on the screen together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 158.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment all the speakers' video feeds are displayed on the screen together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 158.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [158.0, 160.0], "base_se_norm": [0.8975213766098022, 0.9215406179428101], "gt_se_raw": [158.0, 159.0], "gt_se_norm": [0.8975213766098022, 0.9095309972763062], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005497.pt"}
{"idx": 5499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i07EkgDQvy8.mp4", "duration": 132.52, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where two children jump out of laundry baskets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where two children jump out of laundry baskets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 33.0], "base_se_norm": [-0.6226984858512878, -0.5019619464874268], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.5472381711006165, -0.5019619464874268], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005499.pt"}
{"idx": 5501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i07EkgDQvy8.mp4", "duration": 132.52, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the baby is drinking milk from a bottle in the basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 89.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the baby is drinking milk from a bottle in the basket.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 89.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 89.0], "base_se_norm": [0.2979172468185425, 0.3431934118270874], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [0.2979172468185425, 0.3431934118270874], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005501.pt"}
{"idx": 5503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6oyQO-H9YxM.mp4", "duration": 107.732625, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where people are lining up at the airport.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where people are lining up at the airport.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 11.0], "base_se_norm": [-0.9257420897483826, -0.7957907319068909], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.9257420897483826, -0.7957907319068909], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005503.pt"}
{"idx": 5505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6oyQO-H9YxM.mp4", "duration": 107.732625, "prompt_text": "Please find the visual event described by the sentence 'Find the animated sequence demonstrating how the seats board the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the animated sequence demonstrating how the seats board the plane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 44.0], "base_se_norm": [-0.4430656433105469, -0.18316298723220825], "gt_se_raw": [39.0, 42.0], "gt_se_norm": [-0.27598536014556885, -0.22029191255569458], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005505.pt"}
{"idx": 5507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6oyQO-H9YxM.mp4", "duration": 107.732625, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the flight attendant reveals that the entire concept is an April Fools' joke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the flight attendant reveals that the entire concept is an April Fools' joke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 78.0], "base_se_norm": [0.37377142906188965, 0.4480292797088623], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [0.3923358917236328, 0.4480292797088623], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005507.pt"}
{"idx": 5509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGm_DqYtm_Q.mp4", "duration": 11.311, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person uses a pink hairbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person uses a pink hairbrush.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.9450093507766724], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.9450093507766724], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005509.pt"}
{"idx": 5511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nGm_DqYtm_Q.mp4", "duration": 11.311, "prompt_text": "Please find the visual event described by the sentence 'At what point can we see the person grooming their long brown hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point can we see the person grooming their long brown hair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, 0.9450093507766724], "gt_se_raw": [0.0, 11.0], "gt_se_norm": [-1.0, 0.9450093507766724], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005511.pt"}
{"idx": 5513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-6rP8FO_Vwc.mp4", "duration": 364.5308333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man drinking water from a paper cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 222.0 - 226.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man drinking water from a paper cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 222.0 - 226.0 seconds.<|im_end|>\n", "base_se_raw": [222.0, 226.0], "base_se_norm": [0.2180039882659912, 0.2399500608444214], "gt_se_raw": [222.0, 225.0], "gt_se_norm": [0.2180039882659912, 0.23446345329284668], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005513.pt"}
{"idx": 5515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-6rP8FO_Vwc.mp4", "duration": 364.5308333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the waiter scrape the toppings off the pizza?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 98.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the waiter scrape the toppings off the pizza?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 98.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [98.0, 106.0], "base_se_norm": [-0.46232253313064575, -0.41843050718307495], "gt_se_raw": [121.0, 127.0], "gt_se_norm": [-0.3361329436302185, -0.3032139539718628], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005515.pt"}
{"idx": 5517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-6rP8FO_Vwc.mp4", "duration": 364.5308333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the waiter wipe the man's nose with a napkin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 179.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the waiter wipe the man's nose with a napkin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 179.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 179.0], "base_se_norm": [-0.050834715366363525, -0.017915666103363037], "gt_se_raw": [173.0, 179.0], "gt_se_norm": [-0.050834715366363525, -0.017915666103363037], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005517.pt"}
{"idx": 5519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/stoYe8VhIcY.mp4", "duration": 49.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the neon pothos plant is being described.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the neon pothos plant is being described.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 23.0], "base_se_norm": [-0.7981157302856445, -0.07133245468139648], "gt_se_raw": [6.0, 20.0], "gt_se_norm": [-0.7577388882637024, -0.1924629807472229], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005519.pt"}
{"idx": 5521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/stoYe8VhIcY.mp4", "duration": 49.53333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the fiddle-leaf fig displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the fiddle-leaf fig displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 48.0], "base_se_norm": [0.5746971368789673, 0.9380888938903809], "gt_se_raw": [39.0, 48.0], "gt_se_norm": [0.5746971368789673, 0.9380888938903809], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005521.pt"}
{"idx": 5523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6TEwKxBs78k.mp4", "duration": 160.3602, "prompt_text": "Please find the visual event described by the sentence 'At what point is the sign with the text 'I'm watching you.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the sign with the text 'I'm watching you.' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.8752807974815369, -0.8129211664199829], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.8378649950027466, -0.8129211664199829], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005523.pt"}
{"idx": 5525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6TEwKxBs78k.mp4", "duration": 160.3602, "prompt_text": "Please find the visual event described by the sentence 'Show me when the word 'prank'd' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the word 'prank'd' appears on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 56.0], "base_se_norm": [-0.32651615142822266, -0.301572322845459], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.32651615142822266, -0.301572322845459], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005525.pt"}
{"idx": 5527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6TEwKxBs78k.mp4", "duration": 160.3602, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the character in white armor is attacked and defeated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the character in white armor is attacked and defeated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 149.0], "base_se_norm": [0.671237587928772, 0.8583164215087891], "gt_se_raw": [145.0, 149.0], "gt_se_norm": [0.8084287643432617, 0.8583164215087891], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005527.pt"}
{"idx": 5529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M1GO1X09Gec.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where a person is inserting an IV into a mannequin's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where a person is inserting an IV into a mannequin's hand.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.4753605127334595, -0.08188092708587646], "gt_se_raw": [7.0, 14.0], "gt_se_norm": [-0.5409404635429382, -0.08188092708587646], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005529.pt"}
{"idx": 5531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M1GO1X09Gec.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman seen using a microscope in a lab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman seen using a microscope in a lab?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005531.pt"}
{"idx": 5533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fCFkOtlE9zg.mp4", "duration": 54.304249999999996, "prompt_text": "Please find the visual event described by the sentence 'When are the characters bailing water out of a sinking boat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the characters bailing water out of a sinking boat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 14.0], "base_se_norm": [-0.8526818752288818, -0.4843865633010864], "gt_se_raw": [5.0, 14.0], "gt_se_norm": [-0.8158523440361023, -0.4843865633010864], "start_token_span": [73, 76], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005533.pt"}
{"idx": 5535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fCFkOtlE9zg.mp4", "duration": 54.304249999999996, "prompt_text": "Please find the visual event described by the sentence 'At what point is the brain character floating in the water wearing a life vest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the brain character floating in the water wearing a life vest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.11609125137329102, 0.1785449981689453], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.11609125137329102, 0.1785449981689453], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005535.pt"}
{"idx": 5537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fCFkOtlE9zg.mp4", "duration": 54.304249999999996, "prompt_text": "Please find the visual event described by the sentence 'When can we see the brain character trying to start a fire on a deserted island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the brain character trying to start a fire on a deserted island?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 48.0], "base_se_norm": [0.6941584348678589, 0.767817497253418], "gt_se_raw": [46.0, 54.0], "gt_se_norm": [0.6941584348678589, 0.9887946844100952], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005537.pt"}
{"idx": 5539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-AFUpW3oNg.mp4", "duration": 47.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point does the camera pan towards a can of liquid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the camera pan towards a can of liquid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.2772365212440491, -0.10717451572418213], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.2772365212440491, -0.14969003200531006], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005539.pt"}
{"idx": 5541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-AFUpW3oNg.mp4", "duration": 47.041666666666664, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the camera focuses on the side of the wooden component.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the camera focuses on the side of the wooden component.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 46.0], "base_se_norm": [0.5730735063552856, 0.9557129144668579], "gt_se_raw": [36.0, 41.0], "gt_se_norm": [0.5305579900741577, 0.7431354522705078], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005541.pt"}
{"idx": 5543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_La7IMssNOA.mp4", "duration": 73.70704037370704, "prompt_text": "Please find the visual event described by the sentence 'At what point are soldiers seen writing on a form at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are soldiers seen writing on a form at a table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 38.0], "base_se_norm": [-0.05029422044754028, 0.031109094619750977], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.05029422044754028, 0.031109094619750977], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005543.pt"}
{"idx": 5545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_La7IMssNOA.mp4", "duration": 73.70704037370704, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a soldier in an orange vest gestures for a car to move forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a soldier in an orange vest gestures for a car to move forward.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.4030420780181885, -0.3216387629508972], "gt_se_raw": [22.0, 25.0], "gt_se_norm": [-0.4030420780181885, -0.3216387629508972], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005545.pt"}
{"idx": 5547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_La7IMssNOA.mp4", "duration": 73.70704037370704, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where a man in a blue suit is giving a speech at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where a man in a blue suit is giving a speech at a podium.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 71.0], "base_se_norm": [0.5466636419296265, 0.9265459775924683], "gt_se_raw": [57.0, 71.0], "gt_se_norm": [0.5466636419296265, 0.9265459775924683], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005547.pt"}
{"idx": 5549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihlN5nf1qew.mp4", "duration": 94.96987499999999, "prompt_text": "Please find the visual event described by the sentence 'At what point does Michael try to tell the joke to Stanley, who is on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Michael try to tell the joke to Stanley, who is on the phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 31.0], "base_se_norm": [-0.4313986301422119, -0.34716135263442993], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.4313986301422119, -0.34716135263442993], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005549.pt"}
{"idx": 5551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihlN5nf1qew.mp4", "duration": 94.96987499999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where Michael finally succeeds in getting someone with the 'updog' joke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where Michael finally succeeds in getting someone with the 'updog' joke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 60.0], "base_se_norm": [0.09508419036865234, 0.26355862617492676], "gt_se_raw": [52.0, 57.0], "gt_se_norm": [0.09508419036865234, 0.20038068294525146], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005551.pt"}
{"idx": 5553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dL1kDLQ8-kg.mp4", "duration": 41.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'When does the snowman build a smaller version of himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the snowman build a smaller version of himself?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.6129032373428345, -0.3709677457809448], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.5645161271095276, -0.3709677457809448], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005553.pt"}
{"idx": 5555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dL1kDLQ8-kg.mp4", "duration": 41.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point does the small snowman start waddling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the small snowman start waddling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [0.016129016876220703, 0.16129040718078613], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [0.016129016876220703, 0.16129040718078613], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005555.pt"}
{"idx": 5557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dL1kDLQ8-kg.mp4", "duration": 41.333333333333336, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the clip where a snowball with a face flies past Olaf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the clip where a snowball with a face flies past Olaf.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.7096773982048035, -0.6129032373428345], "gt_se_raw": [30.0, 31.0], "gt_se_norm": [0.4516129493713379, 0.5], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005557.pt"}
{"idx": 5559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T4aWsD3jW1w.mp4", "duration": 68.70196666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man let the boy feel the vibration of the violin with his chin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man let the boy feel the vibration of the violin with his chin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 37.0], "base_se_norm": [-0.5342199802398682, 0.07711625099182129], "gt_se_raw": [17.0, 37.0], "gt_se_norm": [-0.5051087737083435, 0.07711625099182129], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005559.pt"}
{"idx": 5561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T4aWsD3jW1w.mp4", "duration": 68.70196666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the boy is eating an ice cream cone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the boy is eating an ice cream cone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 68.0], "base_se_norm": [0.39733994007110596, 0.979564905166626], "gt_se_raw": [48.0, 59.0], "gt_se_norm": [0.39733994007110596, 0.7175636291503906], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005561.pt"}
{"idx": 5563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMb4fgiBimU.mp4", "duration": 81.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the camera panning across the vast field of white tombstones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the camera panning across the vast field of white tombstones?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 26.0], "base_se_norm": [-0.7531880140304565, -0.3582887649536133], "gt_se_raw": [10.0, 25.0], "gt_se_norm": [-0.7531880140304565, -0.3829699754714966], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005563.pt"}
{"idx": 5565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMb4fgiBimU.mp4", "duration": 81.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a bouquet of red roses is placed in front of a grave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a bouquet of red roses is placed in front of a grave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.7285068035125732, -0.6544631719589233], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.3829699754714966, -0.3089263439178467], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005565.pt"}
{"idx": 5567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMb4fgiBimU.mp4", "duration": 81.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man, wearing a red striped shirt and a backpack, is holding a white cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man, wearing a red striped shirt and a backpack, is holding a white cane.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 72.0], "base_se_norm": [-0.16083914041519165, 0.7770464420318604], "gt_se_raw": [73.0, 77.0], "gt_se_norm": [0.8017276525497437, 0.9004524946212769], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005567.pt"}
{"idx": 5569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cWX9iR_PzIQ.mp4", "duration": 193.52686019352686, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the kids are going up the escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the kids are going up the escalator.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 36.0], "base_se_norm": [-0.6899654865264893, -0.6279586553573608], "gt_se_raw": [30.0, 35.0], "gt_se_norm": [-0.6899654865264893, -0.6382930874824524], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005569.pt"}
{"idx": 5571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cWX9iR_PzIQ.mp4", "duration": 193.52686019352686, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the group sneaks into the movie theater and finds their seats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the group sneaks into the movie theater and finds their seats?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 94.0], "base_se_norm": [-0.20424479246139526, -0.02855861186981201], "gt_se_raw": [91.0, 95.0], "gt_se_norm": [-0.05956202745437622, -0.018224120140075684], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005571.pt"}
{"idx": 5573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6R-H-pdufJg.mp4", "duration": 91.05772439105773, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a woman with gray hair is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a woman with gray hair is speaking to the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 19.0], "base_se_norm": [-0.9560717940330505, -0.5826823115348816], "gt_se_raw": [2.0, 20.0], "gt_se_norm": [-0.9560717940330505, -0.5607181787490845], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005573.pt"}
{"idx": 5575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6R-H-pdufJg.mp4", "duration": 91.05772439105773, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman giving a presentation about photochemical smog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman giving a presentation about photochemical smog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 63.0], "base_se_norm": [0.0762404203414917, 0.3837376832962036], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [0.054276347160339355, 0.14213263988494873], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005575.pt"}
{"idx": 5577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6R-H-pdufJg.mp4", "duration": 91.05772439105773, "prompt_text": "Please find the visual event described by the sentence 'When are two men shown working by the water at a treatment plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men shown working by the water at a treatment plant?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 75.0], "base_se_norm": [0.6033785343170166, 0.6473066806793213], "gt_se_raw": [73.0, 75.0], "gt_se_norm": [0.6033785343170166, 0.6473066806793213], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005577.pt"}
{"idx": 5579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bp1kg2jFvJM.mp4", "duration": 162.295, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a hat holds a glass of beer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a hat holds a glass of beer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 42.0], "base_se_norm": [-0.5810098648071289, -0.48242396116256714], "gt_se_raw": [35.0, 39.0], "gt_se_norm": [-0.5686866044998169, -0.5193936824798584], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005579.pt"}
{"idx": 5581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bp1kg2jFvJM.mp4", "duration": 162.295, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman helps the blind woman cross the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman helps the blind woman cross the street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 102.0], "base_se_norm": [0.15838444232940674, 0.2569704055786133], "gt_se_raw": [101.0, 113.0], "gt_se_norm": [0.24464714527130127, 0.3925260305404663], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005581.pt"}
{"idx": 5583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ujncmH8YnNc.mp4", "duration": 188.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the elephant use its trunk to make ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the elephant use its trunk to make ice cream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9682469367980957], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, -0.8941565752029419], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005583.pt"}
{"idx": 5585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ujncmH8YnNc.mp4", "duration": 188.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the little fox dressed as an elephant make a trumpet sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the little fox dressed as an elephant make a trumpet sound?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.5025358200073242, -0.44961410760879517], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.42844539880752563, -0.4072767496109009], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005585.pt"}
{"idx": 5587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ujncmH8YnNc.mp4", "duration": 188.95833333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the rabbit officer pay for the popsicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the rabbit officer pay for the popsicle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 147.0], "base_se_norm": [0.5241456031799316, 0.5558986663818359], "gt_se_raw": [144.0, 146.0], "gt_se_norm": [0.5241456031799316, 0.5453143119812012], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005587.pt"}
{"idx": 5589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AkKEtjlk6Qs.mp4", "duration": 117.95128461795129, "prompt_text": "Please find the visual event described by the sentence 'At what point are sensors shown attached to a person's legs while they are seated on a green chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are sensors shown attached to a person's legs while they are seated on a green chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 51.0], "base_se_norm": [-0.1691485047340393, -0.1352362036705017], "gt_se_raw": [49.0, 52.0], "gt_se_norm": [-0.1691485047340393, -0.11828005313873291], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005589.pt"}
{"idx": 5591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AkKEtjlk6Qs.mp4", "duration": 117.95128461795129, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman walking on the treadmill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman walking on the treadmill?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 88.0], "base_se_norm": [0.45822906494140625, 0.49214136600494385], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [0.45822906494140625, 0.5090975761413574], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005591.pt"}
{"idx": 5593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNV63n3td1A.mp4", "duration": 22.022000000000002, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man and a boy walking down a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man and a boy walking down a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 21.0], "base_se_norm": [-1.0, 0.9071837663650513], "gt_se_raw": [0.0, 21.0], "gt_se_norm": [-1.0, 0.9071837663650513], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005593.pt"}
{"idx": 5595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNV63n3td1A.mp4", "duration": 22.022000000000002, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the boy stops to touch the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the boy stops to touch the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 16.0], "base_se_norm": [-0.2734537720680237, 0.45309245586395264], "gt_se_raw": [10.0, 15.0], "gt_se_norm": [-0.09181725978851318, 0.362274169921875], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005595.pt"}
{"idx": 5597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/cNV63n3td1A.mp4", "duration": 22.022000000000002, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the camera is following the man and the boy from behind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the camera is following the man and the boy from behind.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 21.0], "base_se_norm": [-0.6367268562316895, 0.9071837663650513], "gt_se_raw": [4.0, 21.0], "gt_se_norm": [-0.6367268562316895, 0.9071837663650513], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005597.pt"}
{"idx": 5599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WdKbSQWyqDM.mp4", "duration": 23.056366666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time does the adult guide the baby's hand to pat the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the adult guide the baby's hand to pat the toy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 10.0], "base_se_norm": [-1.0, -0.1325606107711792], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.7397681474685669, -0.5662803053855896], "start_token_span": [76, 79], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005599.pt"}
{"idx": 5601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WdKbSQWyqDM.mp4", "duration": 23.056366666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time frame where the adult's hand is open and gesturing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time frame where the adult's hand is open and gesturing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.21441519260406494], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.21930456161499023, -0.045816659927368164], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005601.pt"}
{"idx": 5603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yb0vU-31FOY.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the boy keeps his face covered with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the boy keeps his face covered with his hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 43.0], "base_se_norm": [-1.0, -0.30992698669433594], "gt_se_raw": [0.0, 43.0], "gt_se_norm": [-1.0, -0.30992698669433594], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005603.pt"}
{"idx": 5605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yb0vU-31FOY.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the boy uncovers his face and starts gesturing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the boy uncovers his face and starts gesturing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 55.0], "base_se_norm": [-0.325975239276886, -0.11734849214553833], "gt_se_raw": [43.0, 47.0], "gt_se_norm": [-0.30992698669433594, -0.24573415517807007], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005605.pt"}
{"idx": 5607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yb0vU-31FOY.mp4", "duration": 124.6245, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the camera pans to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the camera pans to the right.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 104.0], "base_se_norm": [0.5085316896438599, 0.6690137386322021], "gt_se_raw": [95.0, 103.0], "gt_se_norm": [0.5245798826217651, 0.6529655456542969], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005607.pt"}
{"idx": 5609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xeOC6bipMqU.mp4", "duration": 94.88645833333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man's expression change to one of wide-eyed surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man's expression change to one of wide-eyed surprise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 35.0], "base_se_norm": [-0.3255096673965454, -0.26227617263793945], "gt_se_raw": [33.0, 35.0], "gt_se_norm": [-0.30443185567855835, -0.26227617263793945], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005609.pt"}
{"idx": 5611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xeOC6bipMqU.mp4", "duration": 94.88645833333332, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the shadows of the man and woman are kissing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the shadows of the man and woman are kissing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 68.0], "base_se_norm": [0.3489806652069092, 0.4332919120788574], "gt_se_raw": [65.0, 68.0], "gt_se_norm": [0.370058536529541, 0.4332919120788574], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005611.pt"}
{"idx": 5613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ldDAt-vPCY0.mp4", "duration": 195.61208333333332, "prompt_text": "Please find the visual event described by the sentence 'When can we see the bacon being fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the bacon being fried in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 110.0], "base_se_norm": [0.02243173122406006, 0.12467491626739502], "gt_se_raw": [100.0, 110.0], "gt_se_norm": [0.02243173122406006, 0.12467491626739502], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005613.pt"}
{"idx": 5615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ldDAt-vPCY0.mp4", "duration": 195.61208333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dough being rolled out with a rolling pin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dough being rolled out with a rolling pin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 71.0], "base_se_norm": [-0.34564369916915894, -0.27407342195510864], "gt_se_raw": [65.0, 71.0], "gt_se_norm": [-0.33541935682296753, -0.27407342195510864], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005615.pt"}
{"idx": 5617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ldDAt-vPCY0.mp4", "duration": 195.61208333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the cooked bacon strips being chopped on the cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 123.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the cooked bacon strips being chopped on the cutting board?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 123.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 123.0], "base_se_norm": [0.1962451934814453, 0.2575911283493042], "gt_se_raw": [118.0, 123.0], "gt_se_norm": [0.20646953582763672, 0.2575911283493042], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005617.pt"}
{"idx": 5619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1GXQLnTalcg.mp4", "duration": 109.943, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man brushes the front of his teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man brushes the front of his teeth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 51.0], "base_se_norm": [-0.38149768114089966, -0.07224655151367188], "gt_se_raw": [34.0, 40.0], "gt_se_norm": [-0.38149768114089966, -0.2723502516746521], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005619.pt"}
{"idx": 5621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1GXQLnTalcg.mp4", "duration": 109.943, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the man rinses his mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the man rinses his mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 61.0], "base_se_norm": [-0.05405527353286743, 0.1096658706665039], "gt_se_raw": [96.0, 99.0], "gt_se_norm": [0.7463594675064087, 0.8009332418441772], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005621.pt"}
{"idx": 5623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4yG8caPPY1Y.mp4", "duration": 139.12, "prompt_text": "Please find the visual event described by the sentence 'At what point does Steven offer a tray of drinks and fruit slices?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Steven offer a tray of drinks and fruit slices?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 90.0], "base_se_norm": [0.23634278774261475, 0.29384708404541016], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [0.23634278774261475, 0.2794710397720337], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005623.pt"}
{"idx": 5625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4yG8caPPY1Y.mp4", "duration": 139.12, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the scene is drenched in a blue, watery light as Pearl recalls a past battle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the scene is drenched in a blue, watery light as Pearl recalls a past battle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 67.0], "base_se_norm": [-0.22369176149368286, -0.036802709102630615], "gt_se_raw": [53.0, 66.0], "gt_se_norm": [-0.23806780576705933, -0.051178812980651855], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005625.pt"}
{"idx": 5627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4yG8caPPY1Y.mp4", "duration": 139.12, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where Connie and Pearl stand back-to-back, holding their swords up in a victory pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where Connie and Pearl stand back-to-back, holding their swords up in a victory pose.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 122.0], "base_se_norm": [0.6388729810714722, 0.753881573677063], "gt_se_raw": [118.0, 121.0], "gt_se_norm": [0.6963772773742676, 0.7395055294036865], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005627.pt"}
{"idx": 5629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ExeH85oCNIs.mp4", "duration": 95.56213333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the time when the woman is talking about what her friend's daughter likes to eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time when the woman is talking about what her friend's daughter likes to eat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 25.0], "base_se_norm": [-0.7488545179367065, -0.4767801761627197], "gt_se_raw": [28.0, 33.0], "gt_se_norm": [-0.41399383544921875, -0.3093498945236206], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005629.pt"}
{"idx": 5631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ExeH85oCNIs.mp4", "duration": 95.56213333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman discusses the idea of filling a trash can with trinkets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman discusses the idea of filling a trash can with trinkets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 25.0], "base_se_norm": [-0.7488545179367065, -0.4767801761627197], "gt_se_raw": [57.0, 63.0], "gt_se_norm": [0.19294118881225586, 0.3185138702392578], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005631.pt"}
{"idx": 5633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1UdI_eoDPKQ.mp4", "duration": 46.88016666666666, "prompt_text": "Please find the visual event described by the sentence 'When are the dogs playing soccer on the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the dogs playing soccer on the field?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 9.0], "base_se_norm": [-0.7440282106399536, -0.6160423159599304], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.7440282106399536, -0.6160423159599304], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005633.pt"}
{"idx": 5635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1UdI_eoDPKQ.mp4", "duration": 46.88016666666666, "prompt_text": "Please find the visual event described by the sentence 'When are all the Paw Patrol vehicles driving together on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the Paw Patrol vehicles driving together on the road?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 23.0], "base_se_norm": [-0.1467607021331787, -0.018774807453155518], "gt_se_raw": [20.0, 23.0], "gt_se_norm": [-0.1467607021331787, -0.018774807453155518], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005635.pt"}
{"idx": 5637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1UdI_eoDPKQ.mp4", "duration": 46.88016666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the six main pup badges are displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the six main pup badges are displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 9.0], "base_se_norm": [-0.7866901755332947, -0.6160423159599304], "gt_se_raw": [5.0, 6.0], "gt_se_norm": [-0.7866901755332947, -0.7440282106399536], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005637.pt"}
{"idx": 5639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p6qbJMBKato.mp4", "duration": 106.04, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video display the rule about being on time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video display the rule about being on time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 23.0], "base_se_norm": [-0.660505473613739, -0.5662014484405518], "gt_se_raw": [19.0, 23.0], "gt_se_norm": [-0.6416446566581726, -0.5662014484405518], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005639.pt"}
{"idx": 5641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/p6qbJMBKato.mp4", "duration": 106.04, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the message 'Bullying will NOT be tolerated'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the message 'Bullying will NOT be tolerated'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 52.0], "base_se_norm": [-0.13240289688110352, -0.019238054752349854], "gt_se_raw": [46.0, 52.0], "gt_se_norm": [-0.13240289688110352, -0.019238054752349854], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005641.pt"}
{"idx": 5643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9yciDzo8fGo.mp4", "duration": 140.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the movie poster for 'The Theory of Everything' presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the movie poster for 'The Theory of Everything' presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 11.0], "base_se_norm": [-0.971624493598938, -0.8439347147941589], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.985812246799469, -0.943248987197876], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005643.pt"}
{"idx": 5645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9yciDzo8fGo.mp4", "duration": 140.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what moment is the text 'Impacts and Changes' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is the text 'Impacts and Changes' displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [-0.16292262077331543, -0.1061716079711914], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.16292262077331543, -0.12035936117172241], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005645.pt"}
{"idx": 5647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9yciDzo8fGo.mp4", "duration": 140.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Where can we see a cutout of a clock with the words 'LONG TIME' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see a cutout of a clock with the words 'LONG TIME' on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 118.0], "base_se_norm": [0.6032159328460693, 0.6741547584533691], "gt_se_raw": [113.0, 116.0], "gt_se_norm": [0.6032159328460693, 0.6457792520523071], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005647.pt"}
{"idx": 5649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kg-EEBIe7Lk.mp4", "duration": 48.506791666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the magazine is lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the magazine is lying on the ground.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.21660453081130981, 0.07201480865478516], "gt_se_raw": [20.0, 26.0], "gt_se_norm": [-0.17537319660186768, 0.07201480865478516], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005649.pt"}
{"idx": 5651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kg-EEBIe7Lk.mp4", "duration": 48.506791666666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when only the pink and blue gradient background is visible before anything else happens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when only the pink and blue gradient background is visible before anything else happens.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, -0.38152986764907837], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9587686657905579, -0.38152986764907837], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005651.pt"}
{"idx": 5653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j3cEnP8VCvg.mp4", "duration": 19.875, "prompt_text": "Please find the visual event described by the sentence 'Show me when the boy appears angry after the robot leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the boy appears angry after the robot leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.09433960914611816, 0.3081761598587036], "gt_se_raw": [9.0, 13.0], "gt_se_norm": [-0.09433960914611816, 0.3081761598587036], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005653.pt"}
{"idx": 5655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j3cEnP8VCvg.mp4", "duration": 19.875, "prompt_text": "Please find the visual event described by the sentence 'When does a flower pop out of the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a flower pop out of the box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [0.3081761598587036, 0.610062837600708], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [0.3081761598587036, 0.5094339847564697], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005655.pt"}
{"idx": 5657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j3cEnP8VCvg.mp4", "duration": 19.875, "prompt_text": "Please find the visual event described by the sentence 'When is the 'Spoiler alert' text visible on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'Spoiler alert' text visible on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 18.0], "base_se_norm": [0.4088050127029419, 0.8113207817077637], "gt_se_raw": [13.0, 19.0], "gt_se_norm": [0.3081761598587036, 0.911949634552002], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005657.pt"}
{"idx": 5659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W-xuEJVEraw.mp4", "duration": 83.3833, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where a satellite map of Brazil is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where a satellite map of Brazil is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 20.0], "base_se_norm": [-0.6402157545089722, -0.5202876329421997], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.6402157545089722, -0.5202876329421997], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005659.pt"}
{"idx": 5661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W-xuEJVEraw.mp4", "duration": 83.3833, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where people are dancing indoors after the lecture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 58.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where people are dancing indoors after the lecture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 58.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 58.0], "base_se_norm": [0.055367231369018555, 0.3911658525466919], "gt_se_raw": [46.0, 59.0], "gt_se_norm": [0.10333847999572754, 0.4151514768600464], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005661.pt"}
{"idx": 5663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sJzSUK6-O2Q.mp4", "duration": 138.37156666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the instructor guide the student's hands by placing their hands over the student's?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the instructor guide the student's hands by placing their hands over the student's?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 12.0], "base_se_norm": [-0.9421846270561218, -0.8265539407730103], "gt_se_raw": [3.0, 10.0], "gt_se_norm": [-0.956638514995575, -0.8554616570472717], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005663.pt"}
{"idx": 5665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sJzSUK6-O2Q.mp4", "duration": 138.37156666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the yellow paper pieces being attached to the red ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the yellow paper pieces being attached to the red ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 59.0], "base_se_norm": [-0.4941157102584839, -0.14722365140914917], "gt_se_raw": [35.0, 43.0], "gt_se_norm": [-0.4941157102584839, -0.3784850239753723], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005665.pt"}
{"idx": 5667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sJzSUK6-O2Q.mp4", "duration": 138.37156666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the completed craft disassembled and the pieces placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the completed craft disassembled and the pieces placed on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 110.0], "base_se_norm": [0.4598374366760254, 0.5899220705032349], "gt_se_raw": [101.0, 111.0], "gt_se_norm": [0.4598374366760254, 0.6043758392333984], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005667.pt"}
{"idx": 5669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P4vcI-ptEBI.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the dog is licking the birthday cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the dog is licking the birthday cake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 133.0], "base_se_norm": [0.33921074867248535, 0.3807365894317627], "gt_se_raw": [130.0, 133.0], "gt_se_norm": [0.3495922088623047, 0.3807365894317627], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005669.pt"}
{"idx": 5671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/P4vcI-ptEBI.mp4", "duration": 192.65079166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dough being cut into bone shapes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dough being cut into bone shapes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 61.0], "base_se_norm": [-0.42901867628097534, -0.3667297959327698], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [-0.418637216091156, -0.3874927759170532], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005671.pt"}
{"idx": 5673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3T0-olhTvtM.mp4", "duration": 178.8787, "prompt_text": "Please find the visual event described by the sentence 'When can we see three people riding on horseback in the snow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see three people riding on horseback in the snow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 9.0], "base_se_norm": [-0.966457724571228, -0.8993731737136841], "gt_se_raw": [3.0, 6.0], "gt_se_norm": [-0.966457724571228, -0.932915449142456], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005673.pt"}
{"idx": 5675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3T0-olhTvtM.mp4", "duration": 178.8787, "prompt_text": "Please find the visual event described by the sentence 'At what point does the female character say 'My apologies'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the female character say 'My apologies'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 76.0], "base_se_norm": [-0.19498515129089355, -0.15026211738586426], "gt_se_raw": [73.0, 74.0], "gt_se_norm": [-0.18380439281463623, -0.1726236343383789], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005675.pt"}
{"idx": 5677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3T0-olhTvtM.mp4", "duration": 178.8787, "prompt_text": "Please find the visual event described by the sentence 'When does the character shoot three bomb arrows that explode against the rock face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 148.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character shoot three bomb arrows that explode against the rock face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 148.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 148.0], "base_se_norm": [0.4982219934463501, 0.6547527313232422], "gt_se_raw": [136.0, 141.0], "gt_se_norm": [0.5205835103988647, 0.5764874219894409], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005677.pt"}
{"idx": 5679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GBYDe-ERdiQ.mp4", "duration": 234.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are the tombow dual brush pens displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the tombow dual brush pens displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 66.0], "base_se_norm": [-0.4874715209007263, -0.4362186789512634], "gt_se_raw": [61.0, 66.0], "gt_se_norm": [-0.47892940044403076, -0.4362186789512634], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005679.pt"}
{"idx": 5681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GBYDe-ERdiQ.mp4", "duration": 234.13333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point are the stabilo point 88 fineliners shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the stabilo point 88 fineliners shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 137.0], "base_se_norm": [0.1019362211227417, 0.1702733039855957], "gt_se_raw": [130.0, 133.0], "gt_se_norm": [0.11047840118408203, 0.13610482215881348], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005681.pt"}
{"idx": 5683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VemMC_GkmSk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When can we see a variety of medicines neatly arranged on white shelves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a variety of medicines neatly arranged on white shelves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005683.pt"}
{"idx": 5685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VemMC_GkmSk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is someone mixing a pink cream with a spatula?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is someone mixing a pink cream with a spatula?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 21.0], "base_se_norm": [-0.08288431167602539, 0.3756735324859619], "gt_se_raw": [14.0, 21.0], "gt_se_norm": [-0.08288431167602539, 0.3756735324859619], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005685.pt"}
{"idx": 5687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VemMC_GkmSk.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'When does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the text 'LENGTH of PROGRAM' appear on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005687.pt"}
{"idx": 5689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G_QWtUFFAFQ.mp4", "duration": 283.256, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the 'Next Bus' application is opened and used.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 138.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the 'Next Bus' application is opened and used.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 138.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 138.0], "base_se_norm": [-0.13858842849731445, -0.025616466999053955], "gt_se_raw": [123.0, 175.0], "gt_se_norm": [-0.13152772188186646, 0.23563134670257568], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005689.pt"}
{"idx": 5691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G_QWtUFFAFQ.mp4", "duration": 283.256, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the man types and sends a text message.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the man types and sends a text message.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 125.0], "base_se_norm": [-0.2939249873161316, -0.11740618944168091], "gt_se_raw": [179.0, 229.0], "gt_se_norm": [0.2638742923736572, 0.6169118881225586], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005691.pt"}
{"idx": 5693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OMTsvsnO3Mk.mp4", "duration": 151.88, "prompt_text": "Please find the visual event described by the sentence 'At what point does the person in the Adidas hoodie say 'You are still here'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the person in the Adidas hoodie say 'You are still here'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 113.0], "base_se_norm": [0.40900707244873047, 0.48801684379577637], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.7761390805244446, -0.7498024702072144], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005693.pt"}
{"idx": 5695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OMTsvsnO3Mk.mp4", "duration": 151.88, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when the woman with long, dark hair in a grey coat says, 'You're a victor.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when the woman with long, dark hair in a grey coat says, 'You're a victor.'', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 81.0], "base_se_norm": [0.013958334922790527, 0.06663155555725098], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [-0.7102975845336914, -0.683961033821106], "start_token_span": [80, 84], "end_token_span": [86, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005695.pt"}
{"idx": 5697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OMTsvsnO3Mk.mp4", "duration": 151.88, "prompt_text": "Please find the visual event described by the sentence 'When does the man in a black coat and white collared shirt first appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in a black coat and white collared shirt first appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 10.0], "base_se_norm": [-0.8946536779403687, -0.8683171272277832], "gt_se_raw": [8.0, 9.0], "gt_se_norm": [-0.8946536779403687, -0.8814854025840759], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005697.pt"}
{"idx": 5699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JtzsCx0P3tI.mp4", "duration": 35.101733333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where she states that the limbic system processes emotion and sensory systems.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where she states that the limbic system processes emotion and sensory systems.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 28.0], "base_se_norm": [-0.25929588079452515, 0.5953627824783325], "gt_se_raw": [6.0, 10.0], "gt_se_norm": [-0.6581365466117859, -0.4302275776863098], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005699.pt"}
{"idx": 5701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JtzsCx0P3tI.mp4", "duration": 35.101733333333335, "prompt_text": "Please find the visual event described by the sentence 'When does the woman mention 'deafblindness'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman mention 'deafblindness'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.4302275776863098, -0.3162730932235718], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.25349926948547363, 0.36745381355285645], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005701.pt"}
{"idx": 5703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VMbC4Fww1P4.mp4", "duration": 85.83574999999999, "prompt_text": "Please find the visual event described by the sentence 'When is the woman stretching the underwear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman stretching the underwear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 29.0], "base_se_norm": [-0.4873930811882019, -0.3242908716201782], "gt_se_raw": [23.0, 29.0], "gt_se_norm": [-0.4640927314758301, -0.3242908716201782], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005703.pt"}
{"idx": 5705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VMbC4Fww1P4.mp4", "duration": 85.83574999999999, "prompt_text": "Please find the visual event described by the sentence 'At what moment does a woman put the underwear into the washing machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does a woman put the underwear into the washing machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 67.0], "base_se_norm": [0.4679199457168579, 0.5611211061477661], "gt_se_raw": [63.0, 67.0], "gt_se_norm": [0.4679199457168579, 0.5611211061477661], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005705.pt"}
{"idx": 5707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VMbC4Fww1P4.mp4", "duration": 85.83574999999999, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman is holding up the washed underwear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman is holding up the washed underwear.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [0.6543223857879639, 0.7475236654281616], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [0.6543223857879639, 0.7475236654281616], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005707.pt"}
{"idx": 5709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1qM0p24SNhc.mp4", "duration": 35.375, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man says he went snowboarding.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man says he went snowboarding.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 14.0], "base_se_norm": [-0.26501768827438354, -0.20848053693771362], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.3215547800064087, -0.20848053693771362], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005709.pt"}
{"idx": 5711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1qM0p24SNhc.mp4", "duration": 35.375, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man says the weather was a little hot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man says the weather was a little hot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.038869261741638184, 0.13074207305908203], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.038869261741638184, 0.13074207305908203], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005711.pt"}
{"idx": 5713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8B8rjjbVApE.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a woman uses a stethoscope to check the dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a woman uses a stethoscope to check the dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005713.pt"}
{"idx": 5715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8B8rjjbVApE.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man in a white coat holding the dog's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man in a white coat holding the dog's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.017376065254211426, 0.3756735324859619], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.017376065254211426, 0.3756735324859619], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005715.pt"}
{"idx": 5717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8B8rjjbVApE.mp4", "duration": 30.5305, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a woman in blue scrubs is petting the dog's chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a woman in blue scrubs is petting the dog's chest?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5414421558380127], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.5414421558380127], "start_token_span": [74, 77], "end_token_span": [79, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005717.pt"}
{"idx": 5719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hc1hzSvynVY.mp4", "duration": 5.866666666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the time frame where a person in a purple hoodie is pulling a green toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time frame where a person in a purple hoodie is pulling a green toy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7045453786849976], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7045453786849976], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005719.pt"}
{"idx": 5721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hc1hzSvynVY.mp4", "duration": 5.866666666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the interval when a child is interacting with an adult.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the interval when a child is interacting with an adult.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 5.0], "base_se_norm": [-1.0, 0.7045453786849976], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, 0.7045453786849976], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005721.pt"}
{"idx": 5723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ApN73TUVMEU.mp4", "duration": 145.6, "prompt_text": "Please find the visual event described by the sentence 'When does the cat play with the toilet paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat play with the toilet paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 33.0], "base_se_norm": [-0.7527472972869873, -0.5467033386230469], "gt_se_raw": [19.0, 27.0], "gt_se_norm": [-0.7390109896659851, -0.6291208863258362], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005723.pt"}
{"idx": 5725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ApN73TUVMEU.mp4", "duration": 145.6, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the cat knocks over the toothbrushes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the cat knocks over the toothbrushes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.7252747416496277, -0.6153846383094788], "gt_se_raw": [79.0, 82.0], "gt_se_norm": [0.08516478538513184, 0.1263735294342041], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005725.pt"}
{"idx": 5727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ApN73TUVMEU.mp4", "duration": 145.6, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man slips on the wet floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 116.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man slips on the wet floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 116.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [116.0, 120.0], "base_se_norm": [0.5934065580368042, 0.6483515501022339], "gt_se_raw": [117.0, 119.0], "gt_se_norm": [0.6071428060531616, 0.6346153020858765], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005727.pt"}
{"idx": 5729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N_yBGQqg7kM.mp4", "duration": 148.89874999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman with glasses explains that pronouns indicate respect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman with glasses explains that pronouns indicate respect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 83.0], "base_se_norm": [-0.03289979696273804, 0.11485159397125244], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.476154088973999, -0.4224262833595276], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005729.pt"}
{"idx": 5731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N_yBGQqg7kM.mp4", "duration": 148.89874999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the woman in the brown vest describes misgendering as an act of disrespect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the woman in the brown vest describes misgendering as an act of disrespect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 118.0], "base_se_norm": [0.49094605445861816, 0.5849697589874268], "gt_se_raw": [111.0, 119.0], "gt_se_norm": [0.49094605445861816, 0.598401665687561], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005731.pt"}
{"idx": 5733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mxc-TvEhWec.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman assisting the boy in the wheelchair down the corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman assisting the boy in the wheelchair down the corridor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 32.0], "base_se_norm": [-0.9435592889785767, 0.8061035871505737], "gt_se_raw": [1.0, 31.0], "gt_se_norm": [-0.9435592889785767, 0.7496628761291504], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005733.pt"}
{"idx": 5735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mxc-TvEhWec.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the woman is wearing a blue t-shirt and the boy is wearing a red one.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the woman is wearing a blue t-shirt and the boy is wearing a red one.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 32.0], "base_se_norm": [-0.9435592889785767, 0.8061035871505737], "gt_se_raw": [1.0, 31.0], "gt_se_norm": [-0.9435592889785767, 0.7496628761291504], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005735.pt"}
{"idx": 5737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mxc-TvEhWec.mp4", "duration": 35.4354, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment they reach the end of the hallway and prepare to turn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment they reach the end of the hallway and prepare to turn.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 32.0], "base_se_norm": [0.4110184907913208, 0.8061035871505737], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [0.5803406238555908, 0.8061035871505737], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005737.pt"}
{"idx": 5739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pRQLM5nJIeE.mp4", "duration": 122.04, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the woman is pointing at the pictures in the book while reading.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the woman is pointing at the pictures in the book while reading.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.8033431768417358, -0.7377908825874329], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.8197312355041504, -0.7705670595169067], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005739.pt"}
{"idx": 5741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pRQLM5nJIeE.mp4", "duration": 122.04, "prompt_text": "Please find the visual event described by the sentence 'When does the woman show the baby the last page of the book with the 'MR. FUNNY' character on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman show the baby the last page of the book with the 'MR. FUNNY' character on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 113.0], "base_se_norm": [0.7371354103088379, 0.8518518209457397], "gt_se_raw": [107.0, 111.0], "gt_se_norm": [0.7535234689712524, 0.8190757036209106], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005741.pt"}
{"idx": 5743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DpN1zNVoI9E.mp4", "duration": 14.429383471797113, "prompt_text": "Please find the visual event described by the sentence 'Show me when the colorful wheel is spinning at its fastest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the colorful wheel is spinning at its fastest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.4455757737159729], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.4455757737159729], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005743.pt"}
{"idx": 5745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DpN1zNVoI9E.mp4", "duration": 14.429383471797113, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the pointer finally come to a rest on the number 60?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the pointer finally come to a rest on the number 60?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [0.38606059551239014, 0.801878809928894], "gt_se_raw": [8.0, 11.0], "gt_se_norm": [0.1088484525680542, 0.5246666669845581], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005745.pt"}
{"idx": 5747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DpN1zNVoI9E.mp4", "duration": 14.429383471797113, "prompt_text": "Please find the visual event described by the sentence 'Find the period where the blue light on the bottom button is flashing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the period where the blue light on the bottom button is flashing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 14.0], "base_se_norm": [-1.0, 0.940484881401062], "gt_se_raw": [8.0, 13.0], "gt_se_norm": [0.1088484525680542, 0.801878809928894], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005747.pt"}
{"idx": 5749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/69XmzLpL_7A.mp4", "duration": 96.93016666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman puts on a blue mask with a white logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman puts on a blue mask with a white logo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 52.0], "base_se_norm": [-0.13339674472808838, 0.07293736934661865], "gt_se_raw": [47.0, 52.0], "gt_se_norm": [-0.030229687690734863, 0.07293736934661865], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005749.pt"}
{"idx": 5751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/69XmzLpL_7A.mp4", "duration": 96.93016666666666, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a woman removes a disposable medical mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a woman removes a disposable medical mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 77.0], "base_se_norm": [0.4856055974960327, 0.5887726545333862], "gt_se_raw": [72.0, 75.0], "gt_se_norm": [0.4856055974960327, 0.5475058555603027], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005751.pt"}
{"idx": 5753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tXKw_rBhOTs.mp4", "duration": 45.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the first time the tabby cat rings the bell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the first time the tabby cat rings the bell.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.4670614004135132, -0.378238320350647], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.4226498603820801, -0.33382678031921387], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005753.pt"}
{"idx": 5755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tXKw_rBhOTs.mp4", "duration": 45.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the tabby cat lick its nose and lips while looking upwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the tabby cat lick its nose and lips while looking upwards?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 8.0], "base_se_norm": [-0.8667653799057007, -0.6447076201438904], "gt_se_raw": [6.0, 8.0], "gt_se_norm": [-0.7335307002067566, -0.6447076201438904], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005755.pt"}
{"idx": 5757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tXKw_rBhOTs.mp4", "duration": 45.03333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the white cat rings the bell twice in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the white cat rings the bell twice in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 20.0], "base_se_norm": [-0.24500370025634766, -0.11176902055740356], "gt_se_raw": [23.0, 26.0], "gt_se_norm": [0.021465539932250977, 0.15470027923583984], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005757.pt"}
{"idx": 5759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5cLisAxT_v8.mp4", "duration": 55.346958333333326, "prompt_text": "Please find the visual event described by the sentence 'Show me when the text 'Our Goal.' appears on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the text 'Our Goal.' appears on a black screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.638643205165863, -0.4941004514694214], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.6747788786888123, -0.602507472038269], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005759.pt"}
{"idx": 5761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5cLisAxT_v8.mp4", "duration": 55.346958333333326, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the speaker says they are ready to begin producing 3D tactile prints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the speaker says they are ready to begin producing 3D tactile prints.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 42.0], "base_se_norm": [0.3731558322906494, 0.5176985263824463], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [0.26474881172180176, 0.48156285285949707], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005761.pt"}
{"idx": 5763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jQHVrDFNmJE.mp4", "duration": 125.75909242575909, "prompt_text": "Please find the visual event described by the sentence 'Where can we see several mannequins wearing different clothes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see several mannequins wearing different clothes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 38.0], "base_se_norm": [-0.45928364992141724, -0.39566993713378906], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.45928364992141724, -0.39566993713378906], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005763.pt"}
{"idx": 5765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jQHVrDFNmJE.mp4", "duration": 125.75909242575909, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man is using a computer to work on a spreadsheet with clothing designs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man is using a computer to work on a spreadsheet with clothing designs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 79.0], "base_se_norm": [-0.14121520519256592, 0.2563704252243042], "gt_se_raw": [54.0, 59.0], "gt_se_norm": [-0.14121520519256592, -0.061698079109191895], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005765.pt"}
{"idx": 5767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jQHVrDFNmJE.mp4", "duration": 125.75909242575909, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for the display of fashion illustrations and design sketches?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for the display of fashion illustrations and design sketches?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 90.0], "base_se_norm": [0.3199840784072876, 0.4313080310821533], "gt_se_raw": [83.0, 91.0], "gt_se_norm": [0.3199840784072876, 0.4472113847732544], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005767.pt"}
{"idx": 5769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFy3jZk1Tko.mp4", "duration": 68.8, "prompt_text": "Please find the visual event described by the sentence 'When can we see the cat first pop its head out of the grass to look at the butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the cat first pop its head out of the grass to look at the butterfly?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.7093023061752319, -0.5930233001708984], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.680232584476471, -0.6220930814743042], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005769.pt"}
{"idx": 5771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFy3jZk1Tko.mp4", "duration": 68.8, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the cat is frantically chasing and jumping after the butterfly in the tall grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the cat is frantically chasing and jumping after the butterfly in the tall grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 43.0], "base_se_norm": [-0.5930233001708984, 0.25], "gt_se_raw": [31.0, 39.0], "gt_se_norm": [-0.09883725643157959, 0.13372087478637695], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005771.pt"}
{"idx": 5773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6PXKSiJchU.mp4", "duration": 113.15470833333333, "prompt_text": "Please find the visual event described by the sentence 'When are the onions and mushrooms being cooked together in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the onions and mushrooms being cooked together in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 43.0], "base_se_norm": [-0.6111518144607544, -0.2399786114692688], "gt_se_raw": [31.0, 37.0], "gt_se_norm": [-0.45207756757736206, -0.34602808952331543], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005773.pt"}
{"idx": 5775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6PXKSiJchU.mp4", "duration": 113.15470833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment where the grated parmesan cheese is mixed into the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment where the grated parmesan cheese is mixed into the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 65.0], "base_se_norm": [0.06049501895904541, 0.14886951446533203], "gt_se_raw": [59.0, 63.0], "gt_se_norm": [0.042820096015930176, 0.11351966857910156], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005775.pt"}
{"idx": 5777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/I6PXKSiJchU.mp4", "duration": 113.15470833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the finished casserole is topped with fried onions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the finished casserole is topped with fried onions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 87.0], "base_se_norm": [0.48469293117523193, 0.5377177000045776], "gt_se_raw": [84.0, 87.0], "gt_se_norm": [0.48469293117523193, 0.5377177000045776], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005777.pt"}
{"idx": 5779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wezZVZXFO3U.mp4", "duration": 158.3, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when two types of round baking pans, one shiny and one dark, are displayed side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when two types of round baking pans, one shiny and one dark, are displayed side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 47.0], "base_se_norm": [-0.5198988914489746, -0.40619081258773804], "gt_se_raw": [38.0, 47.0], "gt_se_norm": [-0.5198988914489746, -0.40619081258773804], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005779.pt"}
{"idx": 5781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wezZVZXFO3U.mp4", "duration": 158.3, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the cream cheese is being beaten with a hand mixer until it's smooth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the cream cheese is being beaten with a hand mixer until it's smooth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 95.0], "base_se_norm": [-0.09033483266830444, 0.20025265216827393], "gt_se_raw": [72.0, 95.0], "gt_se_norm": [-0.09033483266830444, 0.20025265216827393], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005781.pt"}
{"idx": 5783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0NPYZyI7V8.mp4", "duration": 127.21041666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man start crossing the wide end of the tie over the narrow end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start crossing the wide end of the tie over the narrow end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 52.0], "base_se_norm": [-0.2925108075141907, -0.18245691061019897], "gt_se_raw": [45.0, 48.0], "gt_se_norm": [-0.2925108075141907, -0.24534481763839722], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005783.pt"}
{"idx": 5785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0NPYZyI7V8.mp4", "duration": 127.21041666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does he slide the wide end through the front loop of the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does he slide the wide end through the front loop of the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [-0.009515106678009033, 0.06909477710723877], "gt_se_raw": [93.0, 95.0], "gt_se_norm": [0.4621443748474121, 0.49358832836151123], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005785.pt"}
{"idx": 5787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/T0NPYZyI7V8.mp4", "duration": 127.21041666666666, "prompt_text": "Please find the visual event described by the sentence 'When does he make the final adjustment to the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does he make the final adjustment to the knot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 124.0], "base_se_norm": [0.6979740858078003, 0.9495258331298828], "gt_se_raw": [112.0, 117.0], "gt_se_norm": [0.7608621120452881, 0.8394719362258911], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005787.pt"}
{"idx": 5789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Alc7AhymB9o.mp4", "duration": 45.71233333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the old website featuring John Brademas is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the old website featuring John Brademas is visible.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.5624812841415405, -0.38747382164001465], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.5624812841415405, -0.4749775528907776], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005789.pt"}
{"idx": 5791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Alc7AhymB9o.mp4", "duration": 45.71233333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the text 'WE HAVE LIFT OFF!' appears on a white background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the text 'WE HAVE LIFT OFF!' appears on a white background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 44.0], "base_se_norm": [0.5750672817230225, 0.9250822067260742], "gt_se_raw": [36.0, 39.0], "gt_se_norm": [0.5750672817230225, 0.7063229084014893], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005791.pt"}
{"idx": 5793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OA1ZRGFKRVI.mp4", "duration": 57.09870833333333, "prompt_text": "Please find the visual event described by the sentence 'When are people sitting on the stairs in an auditorium setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people sitting on the stairs in an auditorium setting?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 21.0], "base_se_norm": [-0.40453994274139404, -0.2644317150115967], "gt_se_raw": [18.0, 21.0], "gt_se_norm": [-0.3695129156112671, -0.2644317150115967], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005793.pt"}
{"idx": 5795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OA1ZRGFKRVI.mp4", "duration": 57.09870833333333, "prompt_text": "Please find the visual event described by the sentence 'When is a woman with a tattooed arm speaking into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a woman with a tattooed arm speaking into a microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 41.0], "base_se_norm": [0.2960013151168823, 0.4361095428466797], "gt_se_raw": [41.0, 42.0], "gt_se_norm": [0.4361095428466797, 0.47113656997680664], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005795.pt"}
{"idx": 5797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OA1ZRGFKRVI.mp4", "duration": 57.09870833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a person taking a photo of a group of women with a smartphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a person taking a photo of a group of women with a smartphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 49.0], "base_se_norm": [0.6462719440460205, 0.7163259983062744], "gt_se_raw": [48.0, 49.0], "gt_se_norm": [0.6812989711761475, 0.7163259983062744], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005797.pt"}
{"idx": 5799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wx-kvJlKvJI.mp4", "duration": 90.08999999999999, "prompt_text": "Please find the visual event described by the sentence 'Show the clip of a woman doing pull-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip of a woman doing pull-ups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 38.0], "base_se_norm": [-0.20079916715621948, -0.15639913082122803], "gt_se_raw": [37.0, 38.0], "gt_se_norm": [-0.17859911918640137, -0.15639913082122803], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005799.pt"}
{"idx": 5801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wx-kvJlKvJI.mp4", "duration": 90.08999999999999, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a man swims in the pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a man swims in the pool.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 75.0], "base_se_norm": [0.6206016540527344, 0.6650017499923706], "gt_se_raw": [73.0, 74.0], "gt_se_norm": [0.6206016540527344, 0.6428017616271973], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005801.pt"}
{"idx": 5803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o_P3sKbWbWQ.mp4", "duration": 166.4, "prompt_text": "Please find the visual event described by the sentence 'When does the person in the blue hoodie get killed by an arrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in the blue hoodie get killed by an arrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.8677884340286255, -0.8317307829856873], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.7956730723381042, -0.771634578704834], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005803.pt"}
{"idx": 5805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o_P3sKbWbWQ.mp4", "duration": 166.4, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when a character wearing a pig mask and a crown appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when a character wearing a pig mask and a crown appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 111.0], "base_se_norm": [0.2980769872665405, 0.33413469791412354], "gt_se_raw": [108.0, 110.0], "gt_se_norm": [0.2980769872665405, 0.322115421295166], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005805.pt"}
{"idx": 5807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/o_P3sKbWbWQ.mp4", "duration": 166.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does the character in the blue hoodie start fighting the character with the pig mask?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the character in the blue hoodie start fighting the character with the pig mask?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 134.0], "base_se_norm": [0.5384615659713745, 0.6105769872665405], "gt_se_raw": [130.0, 133.0], "gt_se_norm": [0.5625, 0.598557710647583], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005807.pt"}
{"idx": 5809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G6fMV1UPzkg.mp4", "duration": 147.04704704704704, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when one of the robots falls onto its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 44.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when one of the robots falls onto its back.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 44.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [44.0, 48.0], "base_se_norm": [-0.40155208110809326, -0.34714770317077637], "gt_se_raw": [45.0, 47.0], "gt_se_norm": [-0.38795095682144165, -0.3607487678527832], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005809.pt"}
{"idx": 5811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/G6fMV1UPzkg.mp4", "duration": 147.04704704704704, "prompt_text": "Please find the visual event described by the sentence 'When can we see the robots jumping out from under piles of leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the robots jumping out from under piles of leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 124.0], "base_se_norm": [0.6321307420730591, 0.686535120010376], "gt_se_raw": [120.0, 124.0], "gt_se_norm": [0.6321307420730591, 0.686535120010376], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005811.pt"}
{"idx": 5813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W3v7-3y5THk.mp4", "duration": 178.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in brown show a blue binder to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in brown show a blue binder to the other woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 7.0], "base_se_norm": [-0.966361403465271, -0.9215099811553955], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.966361403465271, -0.9215099811553955], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005813.pt"}
{"idx": 5815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W3v7-3y5THk.mp4", "duration": 178.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the woman in the grey shirt applying a label to the brown box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman in the grey shirt applying a label to the brown box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 73.0], "base_se_norm": [-0.5402728319168091, -0.1814613938331604], "gt_se_raw": [67.0, 76.0], "gt_se_norm": [-0.2487385869026184, -0.14782285690307617], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005815.pt"}
{"idx": 5817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W3v7-3y5THk.mp4", "duration": 178.36666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the woman in the grey shirt opens a new package.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the woman in the grey shirt opens a new package.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 52.0], "base_se_norm": [-0.5402728319168091, -0.4169314503669739], "gt_se_raw": [114.0, 121.0], "gt_se_norm": [0.27826571464538574, 0.35675573348999023], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005817.pt"}
{"idx": 5819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bPa0TiCuvR4.mp4", "duration": 143.30983333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the cricket escapes from the cage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the cricket escapes from the cage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 22.0], "base_se_norm": [-0.7348402738571167, -0.6929728984832764], "gt_se_raw": [19.0, 21.0], "gt_se_norm": [-0.7348402738571167, -0.7069287300109863], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005819.pt"}
{"idx": 5821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/bPa0TiCuvR4.mp4", "duration": 143.30983333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the matchmaker's dress catches fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the matchmaker's dress catches fire.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 110.0], "base_se_norm": [0.47931230068206787, 0.5351355075836182], "gt_se_raw": [99.0, 109.0], "gt_se_norm": [0.38162195682525635, 0.5211796760559082], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005821.pt"}
{"idx": 5823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uasqlGvQhaM.mp4", "duration": 29.4, "prompt_text": "Please find the visual event described by the sentence 'At what point are multiple hands placed on the red container to guide the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are multiple hands placed on the red container to guide the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, -0.2517006993293762], "gt_se_raw": [0.0, 10.0], "gt_se_norm": [-1.0, -0.31972789764404297], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005823.pt"}
{"idx": 5825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uasqlGvQhaM.mp4", "duration": 29.4, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy successfully gets the food out of the container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy successfully gets the food out of the container.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 12.0], "base_se_norm": [-0.31972789764404297, -0.1836734414100647], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.31972789764404297, -0.1836734414100647], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005825.pt"}
{"idx": 5827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uasqlGvQhaM.mp4", "duration": 29.4, "prompt_text": "Please find the visual event described by the sentence 'When can we see the boy eating the snack he just retrieved?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the boy eating the snack he just retrieved?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 18.0], "base_se_norm": [-0.31972789764404297, 0.22448980808258057], "gt_se_raw": [12.0, 20.0], "gt_se_norm": [-0.1836734414100647, 0.36054420471191406], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005827.pt"}
{"idx": 5829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8KveZMqZU6M.mp4", "duration": 149.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man and a woman are running on a dark street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man and a woman are running on a dark street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 45.0], "base_se_norm": [-0.43624162673950195, -0.3959731459617615], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [-0.43624162673950195, -0.38255035877227783], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005829.pt"}
{"idx": 5831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8KveZMqZU6M.mp4", "duration": 149.0, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where one man disintegrates into golden sparks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where one man disintegrates into golden sparks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 93.0], "base_se_norm": [0.18120801448822021, 0.24832212924957275], "gt_se_raw": [87.0, 91.0], "gt_se_norm": [0.16778528690338135, 0.22147655487060547], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005831.pt"}
{"idx": 5833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Cdpf1Dl5b_4.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a pink shirt holding a stack of files?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a pink shirt holding a stack of files?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.8001998066902161, -0.6669996976852417], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.8001998066902161, -0.6669996976852417], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005833.pt"}
{"idx": 5835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Cdpf1Dl5b_4.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is a woman holding a baby with a man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a woman holding a baby with a man's face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 18.0], "base_se_norm": [0.06560099124908447, 0.1988011598587036], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [0.13220107555389404, 0.2654012441635132], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005835.pt"}
{"idx": 5837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Cdpf1Dl5b_4.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'When does a man wearing sunglasses and red and black pants appear from a picnic basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a man wearing sunglasses and red and black pants appear from a picnic basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 24.0], "base_se_norm": [0.46520137786865234, 0.5984015464782715], "gt_se_raw": [22.0, 24.0], "gt_se_norm": [0.46520137786865234, 0.5984015464782715], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005837.pt"}
{"idx": 5839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CPf5cwGYsWs.mp4", "duration": 149.2, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the taxi driver opens the car door for the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the taxi driver opens the car door for the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 54.0], "base_se_norm": [-0.3297587037086487, -0.27613937854766846], "gt_se_raw": [50.0, 53.0], "gt_se_norm": [-0.3297587037086487, -0.2895442247390747], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005839.pt"}
{"idx": 5841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CPf5cwGYsWs.mp4", "duration": 149.2, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a woman is putting a harness on a black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a woman is putting a harness on a black dog.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 104.0], "base_se_norm": [0.34048259258270264, 0.3941018581390381], "gt_se_raw": [100.0, 104.0], "gt_se_norm": [0.34048259258270264, 0.3941018581390381], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005841.pt"}
{"idx": 5843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q3wJ32w4s_A.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman adjusting an X-ray machine over a training dummy's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman adjusting an X-ray machine over a training dummy's head?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.5409404635429382], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.4753605127334595], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005843.pt"}
{"idx": 5845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q3wJ32w4s_A.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point is a teacher giving a lecture to a class of students?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a teacher giving a lecture to a class of students?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.016300976276397705, 0.3771786689758301], "gt_se_raw": [15.0, 22.0], "gt_se_norm": [-0.016300976276397705, 0.44275856018066406], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005845.pt"}
{"idx": 5847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/q3wJ32w4s_A.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a technician is positioning the machine over a patient lying on a bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a technician is positioning the machine over a patient lying on a bed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 28.0], "base_se_norm": [0.44275856018066406, 0.8362381458282471], "gt_se_raw": [22.0, 28.0], "gt_se_norm": [0.44275856018066406, 0.8362381458282471], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005847.pt"}
{"idx": 5849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9C6gDS9YGAU.mp4", "duration": 140.6405, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the basketball is successfully shot through the hoop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the basketball is successfully shot through the hoop.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 37.0], "base_se_norm": [-0.5164977312088013, -0.4738357663154602], "gt_se_raw": [34.0, 36.0], "gt_se_norm": [-0.5164977312088013, -0.4880564212799072], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005849.pt"}
{"idx": 5851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9C6gDS9YGAU.mp4", "duration": 140.6405, "prompt_text": "Please find the visual event described by the sentence 'When does a player in a black jersey fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a player in a black jersey fall to the ground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 92.0], "base_se_norm": [0.26563823223114014, 0.308300256729126], "gt_se_raw": [89.0, 91.0], "gt_se_norm": [0.26563823223114014, 0.2940795421600342], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005851.pt"}
{"idx": 5853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u0p_dBCEDs4.mp4", "duration": 140.30683333333334, "prompt_text": "Please find the visual event described by the sentence 'When is salt being added to the boiling water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is salt being added to the boiling water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.415566623210907, -0.3728032112121582], "gt_se_raw": [41.0, 44.0], "gt_se_norm": [-0.415566623210907, -0.3728032112121582], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005853.pt"}
{"idx": 5855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u0p_dBCEDs4.mp4", "duration": 140.30683333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point are the cooked noodles drained into a colander?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the cooked noodles drained into a colander?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 91.0], "base_se_norm": [0.2116302251815796, 0.29715704917907715], "gt_se_raw": [85.0, 91.0], "gt_se_norm": [0.2116302251815796, 0.29715704917907715], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005855.pt"}
{"idx": 5857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u0p_dBCEDs4.mp4", "duration": 140.30683333333334, "prompt_text": "Please find the visual event described by the sentence 'When are the noodles and sauce being mixed together with tongs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the noodles and sauce being mixed together with tongs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 131.0], "base_se_norm": [0.8245724439620972, 0.8673359155654907], "gt_se_raw": [128.0, 131.0], "gt_se_norm": [0.8245724439620972, 0.8673359155654907], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005857.pt"}
{"idx": 5859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_U-J9PqgmIc.mp4", "duration": 141.07426666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the first time the teacher counts down to stop the music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the first time the teacher counts down to stop the music.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 10.0], "base_se_norm": [-0.9858230948448181, -0.858230710029602], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.47545355558395386, -0.390392005443573], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005859.pt"}
{"idx": 5861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_U-J9PqgmIc.mp4", "duration": 141.07426666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the child chooses a card and says 'ow'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the child chooses a card and says 'ow'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 30.0], "base_se_norm": [-0.6172229051589966, -0.5746921300888062], "gt_se_raw": [10.0, 12.0], "gt_se_norm": [-0.858230710029602, -0.8298768401145935], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005861.pt"}
{"idx": 5863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dZiJkicepzM.mp4", "duration": 160.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the cat hiss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the cat hiss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.9126455783843994, -0.8752079606056213], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.9126455783843994, -0.8876872062683105], "start_token_span": [61, 64], "end_token_span": [66, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005863.pt"}
{"idx": 5865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dZiJkicepzM.mp4", "duration": 160.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show me when a small orange kitten is nuzzling the larger cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a small orange kitten is nuzzling the larger cat.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 98.0], "base_se_norm": [-0.06405985355377197, 0.22296178340911865], "gt_se_raw": [75.0, 86.0], "gt_se_norm": [-0.06405985355377197, 0.07321131229400635], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005865.pt"}
{"idx": 5867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dZiJkicepzM.mp4", "duration": 160.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a tiny black kitten approaches the older cat under a piece of furniture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a tiny black kitten approaches the older cat under a piece of furniture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 74.0], "base_se_norm": [-0.3136439323425293, -0.07653909921646118], "gt_se_raw": [55.0, 64.0], "gt_se_norm": [-0.3136439323425293, -0.20133107900619507], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005867.pt"}
{"idx": 5869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zW39WTnHCc8.mp4", "duration": 82.16541666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when a pink ball appears on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a pink ball appears on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9269765615463257, -0.8539531230926514], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9269765615463257, -0.8782942891120911], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005869.pt"}
{"idx": 5871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zW39WTnHCc8.mp4", "duration": 82.16541666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man start looking from side to side in confusion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man start looking from side to side in confusion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 43.0], "base_se_norm": [-0.2454245686531067, 0.0466691255569458], "gt_se_raw": [31.0, 37.0], "gt_se_norm": [-0.2454245686531067, -0.09937775135040283], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005871.pt"}
{"idx": 5873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZsnaOQwOlI.mp4", "duration": 236.02745833333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman in a pink top come down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman in a pink top come down the stairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 5.0], "base_se_norm": [-0.9745792150497437, -0.9576320648193359], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9745792150497437, -0.9576320648193359], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005873.pt"}
{"idx": 5875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZsnaOQwOlI.mp4", "duration": 236.02745833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman, in a pink sweater, start to panic behind the railing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 61.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman, in a pink sweater, start to panic behind the railing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 61.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 61.0], "base_se_norm": [-0.6695299744606018, -0.48311102390289307], "gt_se_raw": [46.0, 72.0], "gt_se_norm": [-0.6102148294448853, -0.3899015188217163], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005875.pt"}
{"idx": 5877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/qZsnaOQwOlI.mp4", "duration": 236.02745833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman pour the contents of the 'HYPO' bottle into the flower pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 224.0 - 232.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman pour the contents of the 'HYPO' bottle into the flower pot?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 224.0 - 232.0 seconds.<|im_end|>\n", "base_se_raw": [224.0, 232.0], "base_se_norm": [0.8980841636657715, 0.9658728837966919], "gt_se_raw": [232.0, 235.0], "gt_se_norm": [0.9658728837966919, 0.9912936687469482], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005877.pt"}
{"idx": 5879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSTNyHkde08.mp4", "duration": 297.422125, "prompt_text": "Please find the visual event described by the sentence 'An elephant and a giraffe are standing side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'An elephant and a giraffe are standing side by side.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.9260310530662537, -0.8991332650184631], "gt_se_raw": [11.0, 14.0], "gt_se_norm": [-0.9260310530662537, -0.9058576822280884], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005879.pt"}
{"idx": 5881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uSTNyHkde08.mp4", "duration": 297.422125, "prompt_text": "Please find the visual event described by the sentence 'Find the scene that visualizes the three types of dung beetles: rollers, tunnelers, and dwellers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene that visualizes the three types of dung beetles: rollers, tunnelers, and dwellers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [-0.48894184827804565, -0.4553195834159851], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.48221737146377563, -0.4553195834159851], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005881.pt"}
{"idx": 5883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZykGz5kF0U.mp4", "duration": 292.04175, "prompt_text": "Please find the visual event described by the sentence 'When does the pear explain the rules of the pizza challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the pear explain the rules of the pizza challenge?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.9178199768066406, -0.8493366241455078], "gt_se_raw": [13.0, 26.0], "gt_se_norm": [-0.9109716415405273, -0.8219432830810547], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005883.pt"}
{"idx": 5885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZykGz5kF0U.mp4", "duration": 292.04175, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the pear adds black beans to its pizza.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the pear adds black beans to its pizza.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [-0.47952646017074585, -0.44528478384017944], "gt_se_raw": [77.0, 82.0], "gt_se_norm": [-0.47267812490463257, -0.43843644857406616], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005885.pt"}
{"idx": 5887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZykGz5kF0U.mp4", "duration": 292.04175, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the orange character flies away propelled by gas.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 270.0 - 273.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the orange character flies away propelled by gas.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 270.0 - 273.0 seconds.<|im_end|>\n", "base_se_raw": [270.0, 273.0], "base_se_norm": [0.849050760269165, 0.8695957660675049], "gt_se_raw": [269.0, 271.0], "gt_se_norm": [0.8422024250030518, 0.8558990955352783], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005887.pt"}
{"idx": 5889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sL3vqlHabck.mp4", "duration": 171.00416666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the girl asks if they can be picked up to go to the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the girl asks if they can be picked up to go to the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 50.0], "base_se_norm": [-0.5204795598983765, -0.41521894931793213], "gt_se_raw": [21.0, 28.0], "gt_se_norm": [-0.7543919682502747, -0.6725226044654846], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005889.pt"}
{"idx": 5891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/sL3vqlHabck.mp4", "duration": 171.00416666666666, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the boy points outside the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 117.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the boy points outside the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 117.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [117.0, 121.0], "base_se_norm": [0.36838769912719727, 0.41517019271850586], "gt_se_raw": [117.0, 119.0], "gt_se_norm": [0.36838769912719727, 0.39177894592285156], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005891.pt"}
{"idx": 5893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJTgEAuCI8Y.mp4", "duration": 59.684625, "prompt_text": "Please find the visual event described by the sentence 'At what point is the animated logo of the helmet shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the animated logo of the helmet shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 59.0], "base_se_norm": [0.8095113039016724, 0.9770586490631104], "gt_se_raw": [54.0, 58.0], "gt_se_norm": [0.8095113039016724, 0.9435491561889648], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005893.pt"}
{"idx": 5895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJTgEAuCI8Y.mp4", "duration": 59.684625, "prompt_text": "Please find the visual event described by the sentence 'When can we see players looking sad on the bench after a loss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see players looking sad on the bench after a loss?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.731924295425415, -0.5978863835334778], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.731924295425415, -0.5978863835334778], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005895.pt"}
{"idx": 5897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BJTgEAuCI8Y.mp4", "duration": 59.684625, "prompt_text": "Please find the visual event described by the sentence 'Find the segment featuring close-up shots of the championship ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment featuring close-up shots of the championship ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 54.0], "base_se_norm": [0.17283141613006592, 0.8095113039016724], "gt_se_raw": [35.0, 48.0], "gt_se_norm": [0.17283141613006592, 0.6084544658660889], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005897.pt"}
{"idx": 5899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FE6BiUi5HsU.mp4", "duration": 127.7276, "prompt_text": "Please find the visual event described by the sentence 'At what point does the player climb up a tree with pink leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the player climb up a tree with pink leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 6.0], "base_se_norm": [-0.9530250430107117, -0.9060500860214233], "gt_se_raw": [3.0, 5.0], "gt_se_norm": [-0.9530250430107117, -0.9217084050178528], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005899.pt"}
{"idx": 5901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FE6BiUi5HsU.mp4", "duration": 127.7276, "prompt_text": "Please find the visual event described by the sentence 'When does the player hold a TNT block and the music becomes tense?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the player hold a TNT block and the music becomes tense?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 28.0], "base_se_norm": [-0.6242002248764038, -0.561566948890686], "gt_se_raw": [21.0, 24.0], "gt_se_norm": [-0.6711752414703369, -0.6242002248764038], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005901.pt"}
{"idx": 5903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2j6PKrsoy6U.mp4", "duration": 11.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'When are both the boy and the girl sleeping peacefully?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 1.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are both the boy and the girl sleeping peacefully?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 1.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 1.0], "base_se_norm": [-1.0, -0.8250728845596313], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8250728845596313], "start_token_span": [65, 68], "end_token_span": [70, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005903.pt"}
{"idx": 5905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2j6PKrsoy6U.mp4", "duration": 11.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl laugh while hiding under the blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl laugh while hiding under the blanket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 7.0], "base_se_norm": [-0.3002915382385254, 0.22448980808258057], "gt_se_raw": [3.0, 7.0], "gt_se_norm": [-0.47521865367889404, 0.22448980808258057], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005905.pt"}
{"idx": 5907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2j6PKrsoy6U.mp4", "duration": 11.433333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the girl say good night to the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl say good night to the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [0.22448980808258057, 0.7492711544036865], "gt_se_raw": [9.0, 10.0], "gt_se_norm": [0.5743440389633179, 0.7492711544036865], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005907.pt"}
{"idx": 5909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AURiyMmwtOU.mp4", "duration": 319.5192, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a group of dolphins are leaping out of the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 149.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a group of dolphins are leaping out of the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 149.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 149.0], "base_se_norm": [-0.09238630533218384, -0.0673486590385437], "gt_se_raw": [169.0, 175.0], "gt_se_norm": [0.057839393615722656, 0.09539580345153809], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005909.pt"}
{"idx": 5911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AURiyMmwtOU.mp4", "duration": 319.5192, "prompt_text": "Please find the visual event described by the sentence 'When can we see a close-up shot of various colorful flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 233.0 - 248.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a close-up shot of various colorful flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 233.0 - 248.0 seconds.<|im_end|>\n", "base_se_raw": [233.0, 248.0], "base_se_norm": [0.4584413766860962, 0.5523324012756348], "gt_se_raw": [234.0, 247.0], "gt_se_norm": [0.46470069885253906, 0.5460729598999023], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005911.pt"}
{"idx": 5913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UPBMG5EYydo.mp4", "duration": 250.8, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the girl is lying on the grass surrounded by yellow flowers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the girl is lying on the grass surrounded by yellow flowers.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 15.0], "base_se_norm": [-0.9122806787490845, -0.880382776260376], "gt_se_raw": [10.0, 15.0], "gt_se_norm": [-0.920255184173584, -0.880382776260376], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005913.pt"}
{"idx": 5915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UPBMG5EYydo.mp4", "duration": 250.8, "prompt_text": "Please find the visual event described by the sentence 'When does the animation show how water and minerals are transported up the stem to the leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the animation show how water and minerals are transported up the stem to the leaves?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 93.0], "base_se_norm": [-0.42583733797073364, -0.2583732008934021], "gt_se_raw": [84.0, 91.0], "gt_se_norm": [-0.3301435708999634, -0.27432215213775635], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005915.pt"}
{"idx": 5917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UPBMG5EYydo.mp4", "duration": 250.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip that contrasts the gas exchange of a plant during the day and at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 194.0 - 212.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip that contrasts the gas exchange of a plant during the day and at night.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 194.0 - 212.0 seconds.<|im_end|>\n", "base_se_raw": [194.0, 212.0], "base_se_norm": [0.5470494031906128, 0.6905901432037354], "gt_se_raw": [194.0, 212.0], "gt_se_norm": [0.5470494031906128, 0.6905901432037354], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005917.pt"}
{"idx": 5919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q_P8WCbhC6s.mp4", "duration": 118.743625, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the woman sitting on the couch and using her laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the woman sitting on the couch and using her laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 71.0], "base_se_norm": [-0.07363444566726685, 0.19585371017456055], "gt_se_raw": [55.0, 68.0], "gt_se_norm": [-0.07363444566726685, 0.14532470703125], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005919.pt"}
{"idx": 5921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Q_P8WCbhC6s.mp4", "duration": 118.743625, "prompt_text": "Please find the visual event described by the sentence 'When does the woman try to escape through the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 96.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman try to escape through the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 96.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [96.0, 100.0], "base_se_norm": [0.6169289350509644, 0.6843010187149048], "gt_se_raw": [97.0, 99.0], "gt_se_norm": [0.6337720155715942, 0.6674579381942749], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005921.pt"}
{"idx": 5923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fWDaRN490BI.mp4", "duration": 166.3662, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Ed Roberts says we had to teach ourselves how to fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Ed Roberts says we had to teach ourselves how to fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 56.0], "base_se_norm": [-0.38689470291137695, -0.3267863392829895], "gt_se_raw": [54.0, 56.0], "gt_se_norm": [-0.3508296608924866, -0.3267863392829895], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005923.pt"}
{"idx": 5925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fWDaRN490BI.mp4", "duration": 166.3662, "prompt_text": "Please find the visual event described by the sentence 'When does the woman emotionally state that it is a civil right to be independent in America?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 43.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman emotionally state that it is a civil right to be independent in America?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 43.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [43.0, 50.0], "base_se_norm": [-0.48306804895401, -0.3989163637161255], "gt_se_raw": [43.0, 50.0], "gt_se_norm": [-0.48306804895401, -0.3989163637161255], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005925.pt"}
{"idx": 5927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fWDaRN490BI.mp4", "duration": 166.3662, "prompt_text": "Please find the visual event described by the sentence 'When are people seen crawling up the steps of a large government building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 133.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people seen crawling up the steps of a large government building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 133.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 133.0], "base_se_norm": [0.5628174543380737, 0.5988825559616089], "gt_se_raw": [132.0, 137.0], "gt_se_norm": [0.5868607759475708, 0.646969199180603], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005927.pt"}
{"idx": 5929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kKhZtOzakXA.mp4", "duration": 107.39895833333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a man is talking in front of shelves of canned food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a man is talking in front of shelves of canned food.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 37.0], "base_se_norm": [-0.3668467402458191, -0.31098026037216187], "gt_se_raw": [34.0, 38.0], "gt_se_norm": [-0.3668467402458191, -0.2923581004142761], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005929.pt"}
{"idx": 5931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kKhZtOzakXA.mp4", "duration": 107.39895833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman with blonde hair playing the guitar and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman with blonde hair playing the guitar and singing?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 55.0], "base_se_norm": [-0.12475872039794922, 0.02421855926513672], "gt_se_raw": [48.0, 55.0], "gt_se_norm": [-0.10613656044006348, 0.02421855926513672], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005931.pt"}
{"idx": 5933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X4hPwKhjJk0.mp4", "duration": 164.03053333333332, "prompt_text": "Please find the visual event described by the sentence 'When does the man start pouring the liquid into the smoking bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man start pouring the liquid into the smoking bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 9.0], "base_se_norm": [-0.9512286186218262, -0.8902643322944641], "gt_se_raw": [4.0, 8.0], "gt_se_norm": [-0.9512286186218262, -0.9024571776390076], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005933.pt"}
{"idx": 5935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X4hPwKhjJk0.mp4", "duration": 164.03053333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man begin putting the finished product into the glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man begin putting the finished product into the glasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 158.0], "base_se_norm": [0.5119136571884155, 0.9264706373214722], "gt_se_raw": [102.0, 140.0], "gt_se_norm": [0.24367094039916992, 0.7069993019104004], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005935.pt"}
{"idx": 5937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/X4hPwKhjJk0.mp4", "duration": 164.03053333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man hands the finished drink to the customer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man hands the finished drink to the customer.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 162.0], "base_se_norm": [0.9386634826660156, 0.975242018699646], "gt_se_raw": [159.0, 162.0], "gt_se_norm": [0.9386634826660156, 0.975242018699646], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005937.pt"}
{"idx": 5939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V9WcxlXAWCM.mp4", "duration": 31.598333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time do we see a woman in a black and green floral top pointing at a recipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do we see a woman in a black and green floral top pointing at a recipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 14.0], "base_se_norm": [-0.24046629667282104, -0.1138772964477539], "gt_se_raw": [12.0, 14.0], "gt_se_norm": [-0.24046629667282104, -0.1138772964477539], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005939.pt"}
{"idx": 5941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V9WcxlXAWCM.mp4", "duration": 31.598333333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a woman in a blue shirt pours melted chocolate into small white cups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a woman in a blue shirt pours melted chocolate into small white cups.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [0.5823619365692139, 0.7722454071044922], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [0.6456564664840698, 0.7722454071044922], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005941.pt"}
{"idx": 5943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_bdAJXsjvk.mp4", "duration": 8.141466666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a baby barn owl sitting in a red cardboard box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a baby barn owl sitting in a red cardboard box?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, 0.7195917367935181], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.9652477502822876], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005943.pt"}
{"idx": 5945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_bdAJXsjvk.mp4", "duration": 8.141466666666666, "prompt_text": "Please find the visual event described by the sentence 'What is the time range where the owlet is looking at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range where the owlet is looking at the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, 0.7195917367935181], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.9652477502822876], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005945.pt"}
{"idx": 5947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B_bdAJXsjvk.mp4", "duration": 8.141466666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time is the animal with large dark eyes displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the animal with large dark eyes displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, 0.7195917367935181], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, 0.9652477502822876], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005947.pt"}
{"idx": 5949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZxcXr7WaGvc.mp4", "duration": 98.80704166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see someone playing a colorful piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see someone playing a colorful piano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 21.0], "base_se_norm": [-0.6154119968414307, -0.5749291181564331], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.45348024368286133, -0.412997305393219], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005949.pt"}
{"idx": 5951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZxcXr7WaGvc.mp4", "duration": 98.80704166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the exterior of the Braille Institute building shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the exterior of the Braille Institute building shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 6.0], "base_se_norm": [-0.9190341234207153, -0.878551185131073], "gt_se_raw": [4.0, 6.0], "gt_se_norm": [-0.9190341234207153, -0.878551185131073], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005951.pt"}
{"idx": 5953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/czTG7JxOruo.mp4", "duration": 151.38456666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me when the smartphone is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the smartphone is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 25.0], "base_se_norm": [-0.6961381435394287, -0.6697153449058533], "gt_se_raw": [24.0, 25.0], "gt_se_norm": [-0.6829267144203186, -0.6697153449058533], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005953.pt"}
{"idx": 5955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/czTG7JxOruo.mp4", "duration": 151.38456666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment showing statistics about Australians with disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment showing statistics about Australians with disabilities.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 66.0], "base_se_norm": [-0.3130078911781311, -0.12804847955703735], "gt_se_raw": [54.0, 60.0], "gt_se_norm": [-0.28658515214920044, -0.2073168158531189], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005955.pt"}
{"idx": 5957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/czTG7JxOruo.mp4", "duration": 151.38456666666667, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the animation about 'User Testing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the animation about 'User Testing'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 130.0], "base_se_norm": [0.6514233350753784, 0.7174801826477051], "gt_se_raw": [125.0, 129.0], "gt_se_norm": [0.6514233350753784, 0.7042688131332397], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005957.pt"}
{"idx": 5959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4qMyGXTcsg.mp4", "duration": 31.298, "prompt_text": "Please find the visual event described by the sentence 'When does the boy put his fingers on his forehead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy put his fingers on his forehead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 16.0], "base_se_norm": [-1.0, 0.022429585456848145], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.4248833656311035, 0.022429585456848145], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005959.pt"}
{"idx": 5961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/W4qMyGXTcsg.mp4", "duration": 31.298, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the woman's fingers are interlaced with the boy's fingers near his face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the woman's fingers are interlaced with the boy's fingers near his face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 28.0], "base_se_norm": [0.15023326873779297, 0.7892516851425171], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.46974241733551025, 0.5975461006164551], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005961.pt"}
{"idx": 5963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_CE8-CTcRPs.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'A man in blue scrubs is practicing intubation on a medical mannequin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 7.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in blue scrubs is practicing intubation on a medical mannequin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 7.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 7.0], "base_se_norm": [-1.0, -0.6270396113395691], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.6270396113395691], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005963.pt"}
{"idx": 5965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_CE8-CTcRPs.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'At what point is the median salary of $43,496 displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the median salary of $43,496 displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.20079916715621948, 0.11888110637664795], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.20079916715621948, 0.11888110637664795], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005965.pt"}
{"idx": 5967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_CE8-CTcRPs.mp4", "duration": 37.5375, "prompt_text": "Please find the visual event described by the sentence 'Two women in blue uniforms are standing beside a medical machine and smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Two women in blue uniforms are standing beside a medical machine and smiling at the camera.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 35.0], "base_se_norm": [0.5451215505599976, 0.8648018836975098], "gt_se_raw": [28.0, 36.0], "gt_se_norm": [0.49184155464172363, 0.9180819988250732], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005967.pt"}
{"idx": 5969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OkBaZLq7gnU.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman is talking to the sales associate to return items.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 55.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman is talking to the sales associate to return items.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 55.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 55.0], "base_se_norm": [-0.6064127683639526, -0.2784233093261719], "gt_se_raw": [30.0, 46.0], "gt_se_norm": [-0.6064127683639526, -0.39649951457977295], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005969.pt"}
{"idx": 5971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OkBaZLq7gnU.mp4", "duration": 152.4439583333333, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman is making a phone call while sitting in a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman is making a phone call while sitting in a chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 100.0], "base_se_norm": [0.11516392230987549, 0.31195759773254395], "gt_se_raw": [84.0, 113.0], "gt_se_norm": [0.10204434394836426, 0.4825119972229004], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005971.pt"}
{"idx": 5973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5SDeLotL7u0.mp4", "duration": 30.1301, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman is taking a picture of a giant lumberjack statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman is taking a picture of a giant lumberjack statue.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 5.0], "base_se_norm": [-0.8672423958778381, -0.668105959892273], "gt_se_raw": [2.0, 3.0], "gt_se_norm": [-0.8672423958778381, -0.8008636236190796], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005973.pt"}
{"idx": 5975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5SDeLotL7u0.mp4", "duration": 30.1301, "prompt_text": "Please find the visual event described by the sentence 'When does a dog appear on a paddleboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a dog appear on a paddleboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.26983314752578735, -0.07069677114486694], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.26983314752578735, -0.1370755434036255], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005975.pt"}
{"idx": 5977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5SDeLotL7u0.mp4", "duration": 30.1301, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man being served a meal at a food truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man being served a meal at a food truck?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 23.0], "base_se_norm": [0.393954873085022, 0.5267125368118286], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [0.393954873085022, 0.5267125368118286], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005977.pt"}
{"idx": 5979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fF8cAIfSbMw.mp4", "duration": 45.1451, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a boy in a striped shirt and glasses appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a boy in a striped shirt and glasses appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 35.0], "base_se_norm": [-0.9556984305381775, 0.550555944442749], "gt_se_raw": [1.0, 8.0], "gt_se_norm": [-0.9556984305381775, -0.6455872058868408], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005979.pt"}
{"idx": 5981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fF8cAIfSbMw.mp4", "duration": 45.1451, "prompt_text": "Please find the visual event described by the sentence 'When is there a close-up view focusing solely on the hands being massaged?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is there a close-up view focusing solely on the hands being massaged?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 44.0], "base_se_norm": [0.550555944442749, 0.9492702484130859], "gt_se_raw": [35.0, 44.0], "gt_se_norm": [0.550555944442749, 0.9492702484130859], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005981.pt"}
{"idx": 5983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ij9EnG6cF40.mp4", "duration": 122.41666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the black jacket arrive and greet the others?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the black jacket arrive and greet the others?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.8692988157272339, -0.7712729573249817], "gt_se_raw": [9.0, 15.0], "gt_se_norm": [-0.8529611825942993, -0.7549353241920471], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005983.pt"}
{"idx": 5985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ij9EnG6cF40.mp4", "duration": 122.41666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the man in the grey shirt crosses his arms and looks at the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the man in the grey shirt crosses his arms and looks at the woman.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 44.0], "base_se_norm": [-0.46085768938064575, -0.2811436057090759], "gt_se_raw": [33.0, 38.0], "gt_se_norm": [-0.46085768938064575, -0.3791694641113281], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005985.pt"}
{"idx": 5987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ij9EnG6cF40.mp4", "duration": 122.41666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point do all three people laugh together at the end of their conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all three people laugh together at the end of their conversation?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 121.0], "base_se_norm": [0.9278421401977539, 0.9768550395965576], "gt_se_raw": [119.0, 122.0], "gt_se_norm": [0.9441797733306885, 0.9931926727294922], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005987.pt"}
{"idx": 5989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8waFw6Y9fFM.mp4", "duration": 205.93906666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the man attempts to put the first two walls of the house together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the man attempts to put the first two walls of the house together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 66.0], "base_se_norm": [-0.3881685733795166, -0.3590337038040161], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.3881685733795166, -0.3687453269958496], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005989.pt"}
{"idx": 5991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8waFw6Y9fFM.mp4", "duration": 205.93906666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man sprinkles colorful candies over the broken house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 159.0 - 165.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man sprinkles colorful candies over the broken house.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 159.0 - 165.0 seconds.<|im_end|>\n", "base_se_raw": [159.0, 165.0], "base_se_norm": [0.5441460609436035, 0.6024156808853149], "gt_se_raw": [160.0, 164.0], "gt_se_norm": [0.553857684135437, 0.5927040576934814], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005991.pt"}
{"idx": 5993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMfVl38EwpI.mp4", "duration": 110.27683333333333, "prompt_text": "Please find the visual event described by the sentence 'Where can we see people walking through an underwater glass tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see people walking through an underwater glass tunnel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9637276530265808], "gt_se_raw": [0.0, 8.0], "gt_se_norm": [-1.0, -0.8549106121063232], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005993.pt"}
{"idx": 5995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMfVl38EwpI.mp4", "duration": 110.27683333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the scuba diver interact with the children through the glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 47.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the scuba diver interact with the children through the glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 47.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 47.0], "base_se_norm": [-0.29268914461135864, -0.14759975671768188], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.29268914461135864, -0.22014445066452026], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005995.pt"}
{"idx": 5997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZMfVl38EwpI.mp4", "duration": 110.27683333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the time-lapse segment where the tank is being filled with water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 97.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time-lapse segment where the tank is being filled with water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 97.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 97.0], "base_se_norm": [0.6322557926177979, 0.7592090368270874], "gt_se_raw": [91.0, 97.0], "gt_se_norm": [0.6503919363021851, 0.7592090368270874], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005997.pt"}
{"idx": 5999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uKPD6tPry6g.mp4", "duration": 76.60986666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the woman wearing a colorful headscarf on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman wearing a colorful headscarf on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 17.0], "base_se_norm": [-0.7128307223320007, -0.5561928749084473], "gt_se_raw": [11.0, 16.0], "gt_se_norm": [-0.7128307223320007, -0.5822992324829102], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00005999.pt"}
{"idx": 6001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uKPD6tPry6g.mp4", "duration": 76.60986666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment someone uses a small light to inspect audio equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment someone uses a small light to inspect audio equipment.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 56.0], "base_se_norm": [0.2792086601257324, 0.461952805519104], "gt_se_raw": [49.0, 55.0], "gt_se_norm": [0.2792086601257324, 0.4358464479446411], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006001.pt"}
{"idx": 6003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aB_s9lw9E-M.mp4", "duration": 194.6945, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where students are marching in formation into the hall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where students are marching in formation into the hall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 15.0], "base_se_norm": [-1.0, -0.8459124565124512], "gt_se_raw": [1.0, 15.0], "gt_se_norm": [-0.9897274971008301, -0.8459124565124512], "start_token_span": [67, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006003.pt"}
{"idx": 6005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aB_s9lw9E-M.mp4", "duration": 194.6945, "prompt_text": "Please find the visual event described by the sentence 'At what point does the young man with glasses confront the headmaster?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the young man with glasses confront the headmaster?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 158.0], "base_se_norm": [0.3456979990005493, 0.6230555772781372], "gt_se_raw": [132.0, 159.0], "gt_se_norm": [0.35597050189971924, 0.6333280801773071], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006005.pt"}
{"idx": 6007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aB_s9lw9E-M.mp4", "duration": 194.6945, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the man in black robes escapes by shattering the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the man in black robes escapes by shattering the window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 185.0], "base_se_norm": [0.8387781381607056, 0.9004131555557251], "gt_se_raw": [180.0, 185.0], "gt_se_norm": [0.8490506410598755, 0.9004131555557251], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006007.pt"}
{"idx": 6009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkUhEpx7E2k.mp4", "duration": 131.42295555555555, "prompt_text": "Please find the visual event described by the sentence 'When is the ham layered onto the cheese?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the ham layered onto the cheese?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 42.0], "base_se_norm": [-0.4825866222381592, -0.3608422875404358], "gt_se_raw": [33.0, 36.0], "gt_se_norm": [-0.4978046417236328, -0.45215052366256714], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006009.pt"}
{"idx": 6011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MkUhEpx7E2k.mp4", "duration": 131.42295555555555, "prompt_text": "Please find the visual event described by the sentence 'At what point is the dish baking in the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 86.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the dish baking in the oven?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 86.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 86.0], "base_se_norm": [0.18700718879699707, 0.30875158309936523], "gt_se_raw": [79.0, 85.0], "gt_se_norm": [0.2022252082824707, 0.2935335636138916], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006011.pt"}
{"idx": 6013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kPm0_jB5EQI.mp4", "duration": 77.11036666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me when the football players are sitting in the front of the van.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 6.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the football players are sitting in the front of the van.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 6.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 6.0], "base_se_norm": [-1.0, -0.8443788886070251], "gt_se_raw": [0.0, 5.0], "gt_se_norm": [-1.0, -0.8703157305717468], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006013.pt"}
{"idx": 6015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kPm0_jB5EQI.mp4", "duration": 77.11036666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point do the football players rush into the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the football players rush into the house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 16.0], "base_se_norm": [-0.6628209352493286, -0.5850104093551636], "gt_se_raw": [14.0, 16.0], "gt_se_norm": [-0.6368840932846069, -0.5850104093551636], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006015.pt"}
{"idx": 6017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kPm0_jB5EQI.mp4", "duration": 77.11036666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the slogan 'WE WANT YOUR NOISE' displayed on the van's rear doors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the slogan 'WE WANT YOUR NOISE' displayed on the van's rear doors?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.3515787720680237, -0.2218945026397705], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.3515787720680237, -0.2478313446044922], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006017.pt"}
{"idx": 6019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Lt8XAfU_vlM.mp4", "duration": 26.026, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval when the boy moves the blue basket to the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval when the boy moves the blue basket to the blue mat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 8.0], "base_se_norm": [-0.5389226078987122, -0.3852301239967346], "gt_se_raw": [5.0, 8.0], "gt_se_norm": [-0.6157687902450562, -0.3852301239967346], "start_token_span": [71, 74], "end_token_span": [76, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006019.pt"}
{"idx": 6021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Lt8XAfU_vlM.mp4", "duration": 26.026, "prompt_text": "Please find the visual event described by the sentence 'Show me the clip where the boy is holding the red board and pulling the attached pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the clip where the boy is holding the red board and pulling the attached pen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.07784521579742432, 0.3832322359085083], "gt_se_raw": [13.0, 19.0], "gt_se_norm": [-0.0009989738464355469, 0.46007847785949707], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006021.pt"}
{"idx": 6023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QEpCsMbMx7w.mp4", "duration": 179.34583333333333, "prompt_text": "Please find the visual event described by the sentence 'When do two men in black leather jackets appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two men in black leather jackets appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 64.0], "base_se_norm": [-1.0, -0.2862951159477234], "gt_se_raw": [0.0, 63.0], "gt_se_norm": [-1.0, -0.2974467873573303], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006023.pt"}
{"idx": 6025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QEpCsMbMx7w.mp4", "duration": 179.34583333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a red bench is shown alone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a red bench is shown alone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 83.0], "base_se_norm": [-0.14132380485534668, -0.07441401481628418], "gt_se_raw": [77.0, 83.0], "gt_se_norm": [-0.14132380485534668, -0.07441401481628418], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006025.pt"}
{"idx": 6027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QEpCsMbMx7w.mp4", "duration": 179.34583333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when two men are sitting together on the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when two men are sitting together on the bench.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 135.0], "base_se_norm": [0.04825401306152344, 0.5054712295532227], "gt_se_raw": [96.0, 124.0], "gt_se_norm": [0.0705573558807373, 0.38280320167541504], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006027.pt"}
{"idx": 6029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i4eADcCnFjo.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the boy indicates that Dylan is in the lunchroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the boy indicates that Dylan is in the lunchroom.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 33.0], "base_se_norm": [-0.20652979612350464, 0.09102153778076172], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.20652979612350464, -0.1073460578918457], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006029.pt"}
{"idx": 6031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/i4eADcCnFjo.mp4", "duration": 60.493766666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the timeframe where the boy's device says 'in the hallway'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the timeframe where the boy's device says 'in the hallway'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 33.0], "base_se_norm": [-0.20652979612350464, 0.09102153778076172], "gt_se_raw": [35.0, 36.0], "gt_se_norm": [0.15714406967163086, 0.19020521640777588], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006031.pt"}
{"idx": 6033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ccgW5CHFg0Y.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman applying lipstick? please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman applying lipstick? please specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [0.1988011598587036, 0.33200132846832275], "gt_se_raw": [18.0, 19.0], "gt_se_norm": [0.1988011598587036, 0.2654012441635132], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006033.pt"}
{"idx": 6035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ccgW5CHFg0Y.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'Show me when a man is pulling out his empty pockets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a man is pulling out his empty pockets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 8.0], "base_se_norm": [-0.7335997819900513, -0.46719950437545776], "gt_se_raw": [7.0, 8.0], "gt_se_norm": [-0.5337995290756226, -0.46719950437545776], "start_token_span": [66, 69], "end_token_span": [71, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006035.pt"}
{"idx": 6037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ccgW5CHFg0Y.mp4", "duration": 30.029999999999998, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where an olive is dropped into a martini glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where an olive is dropped into a martini glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 26.0], "base_se_norm": [0.665001630783081, 0.7316017150878906], "gt_se_raw": [25.0, 26.0], "gt_se_norm": [0.665001630783081, 0.7316017150878906], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006037.pt"}
{"idx": 6039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vRnhEjP3R-c.mp4", "duration": 162.16199999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the woman in the white collared shirt playing the clarinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the woman in the white collared shirt playing the clarinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 52.0], "base_se_norm": [-0.39566606283187866, -0.3586660623550415], "gt_se_raw": [50.0, 51.0], "gt_se_norm": [-0.3833327293395996, -0.37099939584732056], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006039.pt"}
{"idx": 6041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vRnhEjP3R-c.mp4", "duration": 162.16199999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man burst through a drywall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 124.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man burst through a drywall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 124.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 124.0], "base_se_norm": [0.504668116569519, 0.5293347835540771], "gt_se_raw": [123.0, 124.0], "gt_se_norm": [0.5170015096664429, 0.5293347835540771], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006041.pt"}
{"idx": 6043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dsf_z4urc4s.mp4", "duration": 130.12, "prompt_text": "Please find the visual event described by the sentence 'When can we see a little boy walking through the library aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a little boy walking through the library aisle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 16.0], "base_se_norm": [-0.8462957143783569, -0.7540731430053711], "gt_se_raw": [10.0, 15.0], "gt_se_norm": [-0.8462957143783569, -0.7694435715675354], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006043.pt"}
{"idx": 6045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dsf_z4urc4s.mp4", "duration": 130.12, "prompt_text": "Please find the visual event described by the sentence 'At what time does the man in uniform talk to the children?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 56.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the man in uniform talk to the children?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 56.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 56.0], "base_se_norm": [-0.20073777437210083, -0.1392560601234436], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [-0.20073777437210083, -0.1392560601234436], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006045.pt"}
{"idx": 6047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dsf_z4urc4s.mp4", "duration": 130.12, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the boy is crouched down, drawing on the ground with chalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the boy is crouched down, drawing on the ground with chalk.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 80.0], "base_se_norm": [0.02981865406036377, 0.22963428497314453], "gt_se_raw": [67.0, 77.0], "gt_se_norm": [0.02981865406036377, 0.18352293968200684], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006047.pt"}
{"idx": 6049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pr4CRc_h-9w.mp4", "duration": 190.19, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the father helps his daughter dry her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 38.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the father helps his daughter dry her hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 38.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [38.0, 45.0], "base_se_norm": [-0.6003996133804321, -0.5267890095710754], "gt_se_raw": [39.0, 43.0], "gt_se_norm": [-0.5898838043212891, -0.5478205680847168], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006049.pt"}
{"idx": 6051, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pr4CRc_h-9w.mp4", "duration": 190.19, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the girl finds the photo in her lunchbox.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the girl finds the photo in her lunchbox.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 106.0], "base_se_norm": [0.0620957612991333, 0.11467480659484863], "gt_se_raw": [100.0, 104.0], "gt_se_norm": [0.051579952239990234, 0.0936431884765625], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006051.pt"}
{"idx": 6053, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CwwZOzZtFAM.mp4", "duration": 88.25483333333332, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a man introduces himself as Connor Long.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a man introduces himself as Connor Long.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.728060245513916, -0.5014437437057495], "gt_se_raw": [13.0, 15.0], "gt_se_norm": [-0.7053985595703125, -0.660075306892395], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006053.pt"}
{"idx": 6055, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CwwZOzZtFAM.mp4", "duration": 88.25483333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the driver incorrectly refuse service and drive away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the driver incorrectly refuse service and drive away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 42.0], "base_se_norm": [-0.11619573831558228, -0.04821079969406128], "gt_se_raw": [38.0, 42.0], "gt_se_norm": [-0.1388574242591858, -0.04821079969406128], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006055.pt"}
{"idx": 6057, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CwwZOzZtFAM.mp4", "duration": 88.25483333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene featuring a dog with a thought bubble.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene featuring a dog with a thought bubble.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 66.0], "base_se_norm": [0.40502214431762695, 0.49566876888275146], "gt_se_raw": [62.0, 66.0], "gt_se_norm": [0.40502214431762695, 0.49566876888275146], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006057.pt"}
{"idx": 6059, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7Fjt-mlIlTY.mp4", "duration": 49.315933333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the adult squeezes the yellow toy repeatedly to make it squeak.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the adult squeezes the yellow toy repeatedly to make it squeak.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 23.0], "base_se_norm": [-0.4727870225906372, -0.06723856925964355], "gt_se_raw": [15.0, 20.0], "gt_se_norm": [-0.3916773200035095, -0.1889030933380127], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006059.pt"}
{"idx": 6061, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7Fjt-mlIlTY.mp4", "duration": 49.315933333333334, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the baby is stirring the whisk and the yellow toy together in the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the baby is stirring the whisk and the yellow toy together in the bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 42.0], "base_se_norm": [0.25720012187957764, 0.7033034563064575], "gt_se_raw": [36.0, 42.0], "gt_se_norm": [0.45997440814971924, 0.7033034563064575], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006061.pt"}
{"idx": 6063, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y7HWlprfaGA.mp4", "duration": 15.666666666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the character with monitor heads wave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character with monitor heads wave?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 3.0], "base_se_norm": [-0.8723404407501221, -0.6170213222503662], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.8723404407501221, -0.6170213222503662], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006063.pt"}
{"idx": 6065, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y7HWlprfaGA.mp4", "duration": 15.666666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what time is the character typing on the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the character typing on the keyboard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.3617021441459656, 0.27659571170806885], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.23404258489608765, 0.14893615245819092], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006065.pt"}
{"idx": 6067, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Y7HWlprfaGA.mp4", "duration": 15.666666666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the character standing still near the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the character standing still near the bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 4.0], "base_se_norm": [-1.0, -0.4893617033958435], "gt_se_raw": [0.0, 1.0], "gt_se_norm": [-1.0, -0.8723404407501221], "start_token_span": [64, 67], "end_token_span": [69, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006067.pt"}
{"idx": 6069, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IWrK7sXK3Vw.mp4", "duration": 211.6116116116116, "prompt_text": "Please find the visual event described by the sentence 'Show me when the bunny is eating a carrot for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the bunny is eating a carrot for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 32.0], "base_se_norm": [-0.7826206088066101, -0.697559118270874], "gt_se_raw": [24.0, 32.0], "gt_se_norm": [-0.7731693387031555, -0.697559118270874], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006069.pt"}
{"idx": 6071, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IWrK7sXK3Vw.mp4", "duration": 211.6116116116116, "prompt_text": "Please find the visual event described by the sentence 'At what moment is the bunny sitting outdoors and cleaning its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is the bunny sitting outdoors and cleaning its face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 112.0], "base_se_norm": [-0.054872334003448486, 0.05854296684265137], "gt_se_raw": [101.0, 107.0], "gt_se_norm": [-0.04542100429534912, 0.011286616325378418], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006071.pt"}
{"idx": 6073, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7a4yAbsPCQg.mp4", "duration": 84.618, "prompt_text": "Please find the visual event described by the sentence 'When does the woman point to the thalamus on the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman point to the thalamus on the brain model?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 62.0], "base_se_norm": [0.20541739463806152, 0.4654092788696289], "gt_se_raw": [51.0, 60.0], "gt_se_norm": [0.20541739463806152, 0.4181380271911621], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006073.pt"}
{"idx": 6075, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7a4yAbsPCQg.mp4", "duration": 84.618, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where she explains that the limbic system likes familiarity and to be comfortable.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 49.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where she explains that the limbic system likes familiarity and to be comfortable.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 49.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 49.0], "base_se_norm": [-0.12548154592514038, 0.15814602375030518], "gt_se_raw": [31.0, 35.0], "gt_se_norm": [-0.26729536056518555, -0.17275279760360718], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006075.pt"}
{"idx": 6077, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7a4yAbsPCQg.mp4", "duration": 84.618, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the speaker mentions information being sent to the hypothalamus and pituitary gland.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the speaker mentions information being sent to the hypothalamus and pituitary gland.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 66.0], "base_se_norm": [0.3945024013519287, 0.559951901435852], "gt_se_raw": [77.0, 82.0], "gt_se_norm": [0.8199437856674194, 0.938122034072876], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006077.pt"}
{"idx": 6079, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hDUsLqVeyy4.mp4", "duration": 75.34193333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the boy becomes very excited after hearing the tablet say 'I like fire trucks' for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 3.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the boy becomes very excited after hearing the tablet say 'I like fire trucks' for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 3.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [3.0, 10.0], "base_se_norm": [-0.9203630685806274, -0.7345435619354248], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.07090252637863159, 0.008734345436096191], "start_token_span": [79, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006079.pt"}
{"idx": 6081, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hDUsLqVeyy4.mp4", "duration": 75.34193333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy laugh heartily and throw his head back after the man in the blue shirt signs to him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy laugh heartily and throw his head back after the man in the blue shirt signs to him?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 45.0], "base_se_norm": [0.08837127685546875, 0.19455385208129883], "gt_se_raw": [53.0, 55.0], "gt_se_norm": [0.406919002532959, 0.460010290145874], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006081.pt"}
{"idx": 6083, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g5Bfey0nuGI.mp4", "duration": 169.29412499999998, "prompt_text": "Please find the visual event described by the sentence 'When can we observe a woman doing crunches on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe a woman doing crunches on the floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 142.0], "base_se_norm": [0.5594161748886108, 0.6775537729263306], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.6810284852981567, -0.6455872058868408], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006083.pt"}
{"idx": 6085, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g5Bfey0nuGI.mp4", "duration": 169.29412499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point do the trainers demonstrate standing crunches?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the trainers demonstrate standing crunches?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 65.0], "base_se_norm": [-0.468380868434906, -0.23210567235946655], "gt_se_raw": [51.0, 64.0], "gt_se_norm": [-0.39749830961227417, -0.24391943216323853], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006085.pt"}
{"idx": 6087, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g5Bfey0nuGI.mp4", "duration": 169.29412499999998, "prompt_text": "Please find the visual event described by the sentence 'When can we see a woman reading a book in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a woman reading a book in bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.6455872058868408, -0.6101459264755249], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.6455872058868408, -0.6101459264755249], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006087.pt"}
{"idx": 6089, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gVt9WTJkk7g.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'When are three women sitting on the stairs and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three women sitting on the stairs and talking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 15.0], "base_se_norm": [-0.5959097146987915, -0.494887113571167], "gt_se_raw": [13.0, 18.0], "gt_se_norm": [-0.5622354745864868, -0.39386457204818726], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006089.pt"}
{"idx": 6091, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gVt9WTJkk7g.mp4", "duration": 59.39266666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point are two students working in a fume hood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are two students working in a fume hood?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 39.0], "base_se_norm": [0.010225772857666016, 0.31329345703125], "gt_se_raw": [29.0, 39.0], "gt_se_norm": [-0.023448467254638672, 0.31329345703125], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006091.pt"}
{"idx": 6093, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yXQb9MHDvnI.mp4", "duration": 66.291, "prompt_text": "Please find the visual event described by the sentence 'When is the woman holding the girl's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman holding the girl's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 11.0], "base_se_norm": [-1.0, -0.6681299209594727], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9094899892807007], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006093.pt"}
{"idx": 6095, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yXQb9MHDvnI.mp4", "duration": 66.291, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the woman is holding the girl as the girl struggles with her shoe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the woman is holding the girl as the girl struggles with her shoe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 48.0], "base_se_norm": [-0.004389762878417969, 0.44816040992736816], "gt_se_raw": [32.0, 39.0], "gt_se_norm": [-0.03455972671508789, 0.17663025856018066], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006095.pt"}
{"idx": 6097, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yXQb9MHDvnI.mp4", "duration": 66.291, "prompt_text": "Please find the visual event described by the sentence 'When can we see the feet of another child in pink shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 17.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the feet of another child in pink shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 17.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 17.0], "base_se_norm": [-0.637959897518158, -0.48710983991622925], "gt_se_raw": [11.0, 17.0], "gt_se_norm": [-0.6681299209594727, -0.48710983991622925], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006097.pt"}
{"idx": 6099, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tmg5hLMC1s.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a 'nearly half' comparison is made.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a 'nearly half' comparison is made.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 14.0], "base_se_norm": [-0.03166300058364868, 0.355671763420105], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [0.06517064571380615, 0.25883805751800537], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006099.pt"}
{"idx": 6101, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6tmg5hLMC1s.mp4", "duration": 20.653966666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the moment referencing the 41.1% and 79.1% data points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment referencing the 41.1% and 79.1% data points.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 20.0], "base_se_norm": [-1.0, 0.9366739988327026], "gt_se_raw": [8.0, 14.0], "gt_se_norm": [-0.2253304123878479, 0.355671763420105], "start_token_span": [81, 84], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006101.pt"}
{"idx": 6103, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IKs5-pkF4qE.mp4", "duration": 141.52, "prompt_text": "Please find the visual event described by the sentence 'When does the video introduce the concept of 'Upper Limb Disabilities'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video introduce the concept of 'Upper Limb Disabilities'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 33.0], "base_se_norm": [-0.8021481037139893, -0.5336348414421082], "gt_se_raw": [14.0, 32.0], "gt_se_norm": [-0.8021481037139893, -0.5477671027183533], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006103.pt"}
{"idx": 6105, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IKs5-pkF4qE.mp4", "duration": 141.52, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video discuss how restricted internet access limits global awareness?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video discuss how restricted internet access limits global awareness?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 71.0], "base_se_norm": [-0.3357829451560974, 0.003391742706298828], "gt_se_raw": [47.0, 59.0], "gt_se_norm": [-0.3357829451560974, -0.16619563102722168], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006105.pt"}
{"idx": 6107, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IKs5-pkF4qE.mp4", "duration": 141.52, "prompt_text": "Please find the visual event described by the sentence 'Where can I find the explanation of how the 'Lucy 4' keyboard's laser technology works?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 103.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find the explanation of how the 'Lucy 4' keyboard's laser technology works?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 103.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [103.0, 121.0], "base_se_norm": [0.4556245803833008, 0.7100056409835815], "gt_se_raw": [103.0, 120.0], "gt_se_norm": [0.4556245803833008, 0.6958733797073364], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006107.pt"}
{"idx": 6109, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7WhJ2L5xUqI.mp4", "duration": 54.58786666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a close-up of a 3D painting of a man's face is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 15.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a close-up of a 3D painting of a man's face is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 15.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 15.0], "base_se_norm": [-0.6336182355880737, -0.4504273533821106], "gt_se_raw": [10.0, 15.0], "gt_se_norm": [-0.6336182355880737, -0.4504273533821106], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006109.pt"}
{"idx": 6111, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7WhJ2L5xUqI.mp4", "duration": 54.58786666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a person is touching the tactile 3D art with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a person is touching the tactile 3D art with their hands.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 32.0], "base_se_norm": [-0.1939600706100464, 0.17242169380187988], "gt_se_raw": [22.0, 31.0], "gt_se_norm": [-0.1939600706100464, 0.13578355312347412], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006111.pt"}
{"idx": 6113, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zaSqDJMX_nc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see the child rubbing their eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the child rubbing their eye?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 16.0], "base_se_norm": [-0.4097805619239807, 0.049278974533081055], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.4097805619239807, 0.049278974533081055], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006113.pt"}
{"idx": 6115, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zaSqDJMX_nc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman guide the child's hand to stir the food with the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman guide the child's hand to stir the food with the spoon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 9.0], "base_se_norm": [-0.8688400983810425, -0.4097805619239807], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [0.508338451385498, 0.6394983530044556], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006115.pt"}
{"idx": 6117, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zaSqDJMX_nc.mp4", "duration": 30.497133333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the woman first take the child's hand to help them eat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman first take the child's hand to help them eat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.4097805619239807], "gt_se_raw": [2.0, 5.0], "gt_se_norm": [-0.8688400983810425, -0.672100305557251], "start_token_span": [69, 72], "end_token_span": [74, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006117.pt"}
{"idx": 6119, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tKB4h9gvmm0.mp4", "duration": 112.0, "prompt_text": "Please find the visual event described by the sentence 'Show me when a potato wearing sunglasses pulls out a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a potato wearing sunglasses pulls out a gun.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 26.0], "base_se_norm": [-0.5892857313156128, -0.5357142686843872], "gt_se_raw": [24.0, 26.0], "gt_se_norm": [-0.5714285373687744, -0.5357142686843872], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006119.pt"}
{"idx": 6121, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tKB4h9gvmm0.mp4", "duration": 112.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does a character vomit a rainbow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a character vomit a rainbow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.2857142686843872, -0.2321428656578064], "gt_se_raw": [40.0, 42.0], "gt_se_norm": [-0.2857142686843872, -0.25], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006121.pt"}
{"idx": 6123, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R3tpOXNNMTc.mp4", "duration": 279.8212083333333, "prompt_text": "Please find the visual event described by the sentence 'At what point is a 3D model of a brain shown on a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a 3D model of a brain shown on a computer screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 147.0], "base_se_norm": [0.036376118659973145, 0.05067098140716553], "gt_se_raw": [145.0, 147.0], "gt_se_norm": [0.036376118659973145, 0.05067098140716553], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006123.pt"}
{"idx": 6125, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R3tpOXNNMTc.mp4", "duration": 279.8212083333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a man in a white uniform is practicing martial arts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 228.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a man in a white uniform is practicing martial arts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 228.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 228.0], "base_se_norm": [0.5938749313354492, 0.6296120882034302], "gt_se_raw": [222.0, 227.0], "gt_se_norm": [0.586727499961853, 0.622464656829834], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006125.pt"}
{"idx": 6127, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/R3tpOXNNMTc.mp4", "duration": 279.8212083333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman in a blue top is walking in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 194.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman in a blue top is walking in the park.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 194.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [194.0, 201.0], "base_se_norm": [0.3865997791290283, 0.4366316795349121], "gt_se_raw": [195.0, 200.0], "gt_se_norm": [0.3937472105026245, 0.4294842481613159], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006127.pt"}
{"idx": 6129, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PJlmYh27MHg.mp4", "duration": 208.7085, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the woman in the pink shirt takes off her sunglasses in shock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the woman in the pink shirt takes off her sunglasses in shock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 32.0], "base_se_norm": [-0.7412658929824829, -0.693352222442627], "gt_se_raw": [27.0, 31.0], "gt_se_norm": [-0.7412658929824829, -0.702934980392456], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006129.pt"}
{"idx": 6131, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/PJlmYh27MHg.mp4", "duration": 208.7085, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the old man in the white hat laughs happily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the old man in the white hat laughs happily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 65.0], "base_se_norm": [-0.40586990118026733, -0.3771216869354248], "gt_se_raw": [62.0, 65.0], "gt_se_norm": [-0.40586990118026733, -0.3771216869354248], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006131.pt"}
{"idx": 6133, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQHxqp2H0qI.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man taking the watch from the Lost and Found?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man taking the watch from the Lost and Found?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 32.0], "base_se_norm": [-0.8400000333786011, -0.7866666316986084], "gt_se_raw": [25.0, 31.0], "gt_se_norm": [-0.8333333134651184, -0.7933333516120911], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006133.pt"}
{"idx": 6135, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQHxqp2H0qI.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'When does the protagonist count the money he stole in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 134.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the protagonist count the money he stole in the car?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 134.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [134.0, 146.0], "base_se_norm": [-0.1066666841506958, -0.026666641235351562], "gt_se_raw": [138.0, 145.0], "gt_se_norm": [-0.07999998331069946, -0.03333336114883423], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006135.pt"}
{"idx": 6137, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OQHxqp2H0qI.mp4", "duration": 300.0, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the man tosses the pocket watch into the lake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 263.0 - 266.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the man tosses the pocket watch into the lake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 263.0 - 266.0 seconds.<|im_end|>\n", "base_se_raw": [263.0, 266.0], "base_se_norm": [0.753333330154419, 0.7733333110809326], "gt_se_raw": [264.0, 266.0], "gt_se_norm": [0.7599999904632568, 0.7733333110809326], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006137.pt"}
{"idx": 6139, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nWjEYlMZnhE.mp4", "duration": 238.4, "prompt_text": "Please find the visual event described by the sentence 'Where can I see the paper being folded into a rectangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I see the paper being folded into a rectangle?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 63.0], "base_se_norm": [-0.5469798445701599, -0.4714764952659607], "gt_se_raw": [53.0, 63.0], "gt_se_norm": [-0.5553691387176514, -0.4714764952659607], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006139.pt"}
{"idx": 6141, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nWjEYlMZnhE.mp4", "duration": 238.4, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the creator is forming the head of the butterfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the creator is forming the head of the butterfly.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 163.0], "base_se_norm": [0.19966447353363037, 0.36744964122772217], "gt_se_raw": [142.0, 163.0], "gt_se_norm": [0.1912752389907837, 0.36744964122772217], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006141.pt"}
{"idx": 6143, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-AKDXFx4jI.mp4", "duration": 201.33446666666666, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a black suit playing the piano by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 35.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a black suit playing the piano by the water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 35.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 35.0], "base_se_norm": [-0.7417232990264893, -0.6523198485374451], "gt_se_raw": [25.0, 34.0], "gt_se_norm": [-0.7516570091247559, -0.6622535586357117], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006143.pt"}
{"idx": 6145, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-AKDXFx4jI.mp4", "duration": 201.33446666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where a man with a name tag 'David' is being interviewed in front of the piano.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where a man with a name tag 'David' is being interviewed in front of the piano.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 83.0], "base_se_norm": [-0.31457340717315674, -0.17550134658813477], "gt_se_raw": [68.0, 83.0], "gt_se_norm": [-0.32450711727142334, -0.17550134658813477], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006145.pt"}
{"idx": 6147, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/k-AKDXFx4jI.mp4", "duration": 201.33446666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a woman in sunglasses plays the piano while a man plays the saxophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a woman in sunglasses plays the piano while a man plays the saxophone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 160.0], "base_se_norm": [0.5397263765335083, 0.5893949270248413], "gt_se_raw": [156.0, 160.0], "gt_se_norm": [0.5496600866317749, 0.5893949270248413], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006147.pt"}
{"idx": 6149, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1cHSTAv6aw.mp4", "duration": 227.39383333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the AT Specialist & Occupational Therapist introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the AT Specialist & Occupational Therapist introduces herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [-0.4810764789581299, -0.42830461263656616], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.4810764789581299, -0.42830461263656616], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006149.pt"}
{"idx": 6151, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_1cHSTAv6aw.mp4", "duration": 227.39383333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy listen to the recording from his smart pen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 129.0 - 135.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy listen to the recording from his smart pen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 129.0 - 135.0 seconds.<|im_end|>\n", "base_se_raw": [129.0, 135.0], "base_se_norm": [0.13459539413452148, 0.18736732006072998], "gt_se_raw": [129.0, 132.0], "gt_se_norm": [0.13459539413452148, 0.1609814167022705], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006151.pt"}
{"idx": 6153, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBF5l9rWQFs.mp4", "duration": 286.625, "prompt_text": "Please find the visual event described by the sentence 'When do the women dressed as villains gather around a candlelit table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the women dressed as villains gather around a candlelit table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 18.0], "base_se_norm": [-0.9651111960411072, -0.8744003772735596], "gt_se_raw": [5.0, 13.0], "gt_se_norm": [-0.9651111960411072, -0.9092891216278076], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006153.pt"}
{"idx": 6155, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBF5l9rWQFs.mp4", "duration": 286.625, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the blue dress starts singing angrily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the blue dress starts singing angrily.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 63.0], "base_se_norm": [-0.6650675535202026, -0.560401201248169], "gt_se_raw": [47.0, 50.0], "gt_se_norm": [-0.6720453500747681, -0.6511120796203613], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006155.pt"}
{"idx": 6157, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kBF5l9rWQFs.mp4", "duration": 286.625, "prompt_text": "Please find the visual event described by the sentence 'When can we see the animated clip of a witch preparing a poison apple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the animated clip of a witch preparing a poison apple?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 117.0], "base_se_norm": [-0.21151328086853027, -0.18360227346420288], "gt_se_raw": [215.0, 217.0], "gt_se_norm": [0.5002180337905884, 0.5141736268997192], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006157.pt"}
{"idx": 6159, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mwm0OwqWvF4.mp4", "duration": 305.56, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment a dog jumps into a car through an open window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment a dog jumps into a car through an open window.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 116.0], "base_se_norm": [-0.28655582666397095, -0.24073833227157593], "gt_se_raw": [108.0, 111.0], "gt_se_norm": [-0.2931011915206909, -0.273465096950531], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006159.pt"}
{"idx": 6161, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mwm0OwqWvF4.mp4", "duration": 305.56, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the large dog and a much smaller dog are playing together on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 229.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the large dog and a much smaller dog are playing together on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 229.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 229.0], "base_se_norm": [0.4596151113510132, 0.498887300491333], "gt_se_raw": [224.0, 229.0], "gt_se_norm": [0.46616053581237793, 0.498887300491333], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006161.pt"}
{"idx": 6163, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ygqARUR7aW8.mp4", "duration": 228.0278, "prompt_text": "Please find the visual event described by the sentence 'When is the parrot spinning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 192.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the parrot spinning?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 192.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [192.0, 198.0], "base_se_norm": [0.6840051412582397, 0.7366303205490112], "gt_se_raw": [191.0, 198.0], "gt_se_norm": [0.675234317779541, 0.7366303205490112], "start_token_span": [61, 66], "end_token_span": [68, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006163.pt"}
{"idx": 6165, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ygqARUR7aW8.mp4", "duration": 228.0278, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the parrot hangs upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 206.0 - 214.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the parrot hangs upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 206.0 - 214.0 seconds.<|im_end|>\n", "base_se_raw": [206.0, 214.0], "base_se_norm": [0.8067972660064697, 0.8769640922546387], "gt_se_raw": [206.0, 215.0], "gt_se_norm": [0.8067972660064697, 0.8857349157333374], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006165.pt"}
{"idx": 6167, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ygqARUR7aW8.mp4", "duration": 228.0278, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the parrot gives a high five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 184.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the parrot gives a high five.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 184.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [184.0, 188.0], "base_se_norm": [0.6138383150100708, 0.6489217281341553], "gt_se_raw": [184.0, 188.0], "gt_se_norm": [0.6138383150100708, 0.6489217281341553], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006167.pt"}
{"idx": 6169, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ijCJg0-nL-U.mp4", "duration": 297.297, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman apply concealer to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 125.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman apply concealer to her face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 125.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 125.0], "base_se_norm": [-0.47527217864990234, -0.1590900421142578], "gt_se_raw": [79.0, 88.0], "gt_se_norm": [-0.4685449004173279, -0.4079993963241577], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006169.pt"}
{"idx": 6171, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ijCJg0-nL-U.mp4", "duration": 297.297, "prompt_text": "Please find the visual event described by the sentence 'Where in the video does she put on the false eyelashes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video does she put on the false eyelashes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 169.0], "base_se_norm": [0.02254652976989746, 0.13691020011901855], "gt_se_raw": [153.0, 160.0], "gt_se_norm": [0.02927374839782715, 0.07636475563049316], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006171.pt"}
{"idx": 6173, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TuE6z8X-rGg.mp4", "duration": 303.7034, "prompt_text": "Please find the visual event described by the sentence 'When is a person collecting eggs from the coop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 53.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person collecting eggs from the coop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 53.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [53.0, 80.0], "base_se_norm": [-0.650975227355957, -0.4731702208518982], "gt_se_raw": [53.0, 78.0], "gt_se_norm": [-0.650975227355957, -0.4863409399986267], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006173.pt"}
{"idx": 6175, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TuE6z8X-rGg.mp4", "duration": 303.7034, "prompt_text": "Please find the visual event described by the sentence 'At what point is a boy using a walker to move down a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 171.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a boy using a walker to move down a hallway?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 171.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 171.0], "base_se_norm": [0.07341575622558594, 0.1260986328125], "gt_se_raw": [200.0, 209.0], "gt_se_norm": [0.3170745372772217, 0.37634289264678955], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006175.pt"}
{"idx": 6177, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TuE6z8X-rGg.mp4", "duration": 303.7034, "prompt_text": "Please find the visual event described by the sentence 'A boy in a wheelchair is using an adaptive tool to check the mailbox. Specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 242.0 - 266.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A boy in a wheelchair is using an adaptive tool to check the mailbox. Specify the time range.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 242.0 - 266.0 seconds.<|im_end|>\n", "base_se_raw": [242.0, 266.0], "base_se_norm": [0.5936601161956787, 0.7517091035842896], "gt_se_raw": [254.0, 266.0], "gt_se_norm": [0.6726845502853394, 0.7517091035842896], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006177.pt"}
{"idx": 6179, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vN1fuDgj-Qk.mp4", "duration": 222.18029166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the woman in the denim jacket and plaid top walks down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the woman in the denim jacket and plaid top walks down the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 136.0], "base_se_norm": [0.188224196434021, 0.22423100471496582], "gt_se_raw": [132.0, 134.0], "gt_se_norm": [0.188224196434021, 0.20622754096984863], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006179.pt"}
{"idx": 6181, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vN1fuDgj-Qk.mp4", "duration": 222.18029166666665, "prompt_text": "Please find the visual event described by the sentence 'When are the black shoes with white patterns being showcased?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 179.0 - 187.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the black shoes with white patterns being showcased?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 179.0 - 187.0 seconds.<|im_end|>\n", "base_se_raw": [179.0, 187.0], "base_se_norm": [0.6113040447235107, 0.6833175420761108], "gt_se_raw": [179.0, 187.0], "gt_se_norm": [0.6113040447235107, 0.6833175420761108], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006181.pt"}
{"idx": 6183, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ekpylglrh2U.mp4", "duration": 313.7134, "prompt_text": "Please find the visual event described by the sentence 'When does the couple discuss flipping a coin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the couple discuss flipping a coin?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 66.0], "base_se_norm": [-0.9171218276023865, -0.5792337656021118], "gt_se_raw": [27.0, 32.0], "gt_se_norm": [-0.8278683423995972, -0.7959921360015869], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006183.pt"}
{"idx": 6185, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ekpylglrh2U.mp4", "duration": 313.7134, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman is going down a yellow slide.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman is going down a yellow slide.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 134.0], "base_se_norm": [-0.1584675908088684, -0.1457170844078064], "gt_se_raw": [133.0, 135.0], "gt_se_norm": [-0.1520923376083374, -0.1393418312072754], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006185.pt"}
{"idx": 6187, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ekpylglrh2U.mp4", "duration": 313.7134, "prompt_text": "Please find the visual event described by the sentence 'When are the group of people taking a picture on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 259.0 - 287.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the group of people taking a picture on the grass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 259.0 - 287.0 seconds.<|im_end|>\n", "base_se_raw": [259.0, 287.0], "base_se_norm": [0.6511886119842529, 0.829695463180542], "gt_se_raw": [260.0, 263.0], "gt_se_norm": [0.6575638055801392, 0.6766895055770874], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006187.pt"}
{"idx": 6189, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QVwXvsrJ6UA.mp4", "duration": 274.73333333333335, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when a mother and son are shown baking together in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 80.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when a mother and son are shown baking together in the kitchen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 80.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 80.0], "base_se_norm": [-0.446736216545105, -0.4176170825958252], "gt_se_raw": [77.0, 80.0], "gt_se_norm": [-0.4394564628601074, -0.4176170825958252], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006189.pt"}
{"idx": 6191, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QVwXvsrJ6UA.mp4", "duration": 274.73333333333335, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when a girl is washing a bowl in the sink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when a girl is washing a bowl in the sink?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [-0.4831351637840271, -0.4540160298347473], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [-0.4831351637840271, -0.4540160298347473], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006191.pt"}
{"idx": 6193, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mO3mMYwKkKs.mp4", "duration": 325.24, "prompt_text": "Please find the visual event described by the sentence 'When are the components of a pin-tumbler lock displayed in an animated diagram?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the components of a pin-tumbler lock displayed in an animated diagram?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 69.0], "base_se_norm": [-0.6125937700271606, -0.5756978988647461], "gt_se_raw": [63.0, 69.0], "gt_se_norm": [-0.6125937700271606, -0.5756978988647461], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006193.pt"}
{"idx": 6195, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mO3mMYwKkKs.mp4", "duration": 325.24, "prompt_text": "Please find the visual event described by the sentence 'At what point are the 'Tension Wrenches' presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 162.0 - 174.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the 'Tension Wrenches' presented?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 162.0 - 174.0 seconds.<|im_end|>\n", "base_se_raw": [162.0, 174.0], "base_se_norm": [-0.003812551498413086, 0.06997907161712646], "gt_se_raw": [163.0, 169.0], "gt_se_norm": [0.002336740493774414, 0.039232611656188965], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006195.pt"}
{"idx": 6197, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mO3mMYwKkKs.mp4", "duration": 325.24, "prompt_text": "Please find the visual event described by the sentence 'When is the 'raking' technique demonstrated on the clear practice lock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 249.0 - 266.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the 'raking' technique demonstrated on the clear practice lock?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 249.0 - 266.0 seconds.<|im_end|>\n", "base_se_raw": [249.0, 266.0], "base_se_norm": [0.5311770439147949, 0.6357152462005615], "gt_se_raw": [251.0, 265.0], "gt_se_norm": [0.5434756278991699, 0.629565954208374], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006197.pt"}
{"idx": 6199, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mvD5wvlIE7c.mp4", "duration": 230.8973333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the team walks towards the In-N-Out Burger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the team walks towards the In-N-Out Burger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 33.0], "base_se_norm": [-0.7401442527770996, -0.7141586542129517], "gt_se_raw": [30.0, 33.0], "gt_se_norm": [-0.7401442527770996, -0.7141586542129517], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006199.pt"}
{"idx": 6201, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mvD5wvlIE7c.mp4", "duration": 230.8973333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when students are giving a presentation to the judges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when students are giving a presentation to the judges.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 116.0], "base_se_norm": [-0.1597997546195984, 0.004775524139404297], "gt_se_raw": [96.0, 103.0], "gt_se_norm": [-0.1684616208076477, -0.10782861709594727], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006201.pt"}
{"idx": 6203, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ezg4sr67OGA.mp4", "duration": 257.67408333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man with long grey hair and a beard walk into the wooden house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man with long grey hair and a beard walk into the wooden house?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 24.0], "base_se_norm": [-0.8835738301277161, -0.8137181401252747], "gt_se_raw": [17.0, 22.0], "gt_se_norm": [-0.8680503368377686, -0.8292416334152222], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006203.pt"}
{"idx": 6205, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ezg4sr67OGA.mp4", "duration": 257.67408333333333, "prompt_text": "Please find the visual event described by the sentence 'What is the time frame where the man with the beard gets shot in the shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 187.0 - 190.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time frame where the man with the beard gets shot in the shoulder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 187.0 - 190.0 seconds.<|im_end|>\n", "base_se_raw": [187.0, 190.0], "base_se_norm": [0.45144593715667725, 0.4747312068939209], "gt_se_raw": [178.0, 181.0], "gt_se_norm": [0.38159024715423584, 0.4048755168914795], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006205.pt"}
{"idx": 6207, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ezg4sr67OGA.mp4", "duration": 257.67408333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the beret get out of his car and walk away from it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the beret get out of his car and walk away from it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 14.0], "base_se_norm": [-0.93790602684021, -0.8913356065750122], "gt_se_raw": [225.0, 229.0], "gt_se_norm": [0.7463922500610352, 0.7774392366409302], "start_token_span": [75, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006207.pt"}
{"idx": 6209, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7N98N0GkGjY.mp4", "duration": 270.10316666666665, "prompt_text": "Please find the visual event described by the sentence 'At what time does the marching band appear on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time does the marching band appear on the street?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 54.0], "base_se_norm": [-0.7260305881500244, -0.6001527905464172], "gt_se_raw": [37.0, 48.0], "gt_se_norm": [-0.7260305881500244, -0.6445802450180054], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006209.pt"}
{"idx": 6211, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7N98N0GkGjY.mp4", "duration": 270.10316666666665, "prompt_text": "Please find the visual event described by the sentence 'A man is painting in a wheatfield, surrounded by crows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 165.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man is painting in a wheatfield, surrounded by crows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 165.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 165.0], "base_se_norm": [0.1254957914352417, 0.22175538539886475], "gt_se_raw": [157.0, 162.0], "gt_se_norm": [0.16251873970031738, 0.19954156875610352], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006211.pt"}
{"idx": 6213, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YXOapWsVmFw.mp4", "duration": 268.47654166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the text 'Rachel Hestres' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the text 'Rachel Hestres' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 18.0], "base_se_norm": [-0.9031572937965393, -0.8659100532531738], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.9031572937965393, -0.8733595013618469], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006213.pt"}
{"idx": 6215, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YXOapWsVmFw.mp4", "duration": 268.47654166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a person is adjusting the controls on an oscilloscope, showing sine waves on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 168.0 - 170.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a person is adjusting the controls on an oscilloscope, showing sine waves on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 168.0 - 170.0 seconds.<|im_end|>\n", "base_se_raw": [168.0, 170.0], "base_se_norm": [0.25150609016418457, 0.2664048671722412], "gt_se_raw": [168.0, 171.0], "gt_se_norm": [0.25150609016418457, 0.2738543748855591], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006215.pt"}
{"idx": 6217, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/YXOapWsVmFw.mp4", "duration": 268.47654166666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the violinist is playing her instrument on a wooden stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 246.0 - 251.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the violinist is playing her instrument on a wooden stage.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 246.0 - 251.0 seconds.<|im_end|>\n", "base_se_raw": [246.0, 251.0], "base_se_norm": [0.8325624465942383, 0.869809627532959], "gt_se_raw": [246.0, 248.0], "gt_se_norm": [0.8325624465942383, 0.8474613428115845], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006217.pt"}
{"idx": 6219, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a4hqZD5g_yQ.mp4", "duration": 231.866, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a woman with pink hair is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a woman with pink hair is talking on the phone.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.792983889579773, -0.7671068906784058], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.7843582034111023, -0.7584812045097351], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006219.pt"}
{"idx": 6221, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a4hqZD5g_yQ.mp4", "duration": 231.866, "prompt_text": "Please find the visual event described by the sentence 'At what point is a man with long, blondish hair and glasses being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a man with long, blondish hair and glasses being interviewed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 68.0], "base_se_norm": [-0.4393313527107239, -0.4134542942047119], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.7239785194396973, -0.6894757747650146], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006221.pt"}
{"idx": 6223, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l1YSzo9n8EE.mp4", "duration": 273.6066666666666, "prompt_text": "Please find the visual event described by the sentence 'When are all the ingredients for the grilled cheese sandwich displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the ingredients for the grilled cheese sandwich displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 32.0], "base_se_norm": [-0.861114501953125, -0.766087532043457], "gt_se_raw": [19.0, 33.0], "gt_se_norm": [-0.861114501953125, -0.7587777972221375], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006223.pt"}
{"idx": 6225, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l1YSzo9n8EE.mp4", "duration": 273.6066666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the sandwich is being flipped in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 174.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the sandwich is being flipped in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 174.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [174.0, 183.0], "base_se_norm": [0.27189886569976807, 0.33768677711486816], "gt_se_raw": [179.0, 183.0], "gt_se_norm": [0.3084477186203003, 0.33768677711486816], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006225.pt"}
{"idx": 6227, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l1YSzo9n8EE.mp4", "duration": 273.6066666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the final plated grilled cheese sandwich is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 235.0 - 243.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the final plated grilled cheese sandwich is presented.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 235.0 - 243.0 seconds.<|im_end|>\n", "base_se_raw": [235.0, 243.0], "base_se_norm": [0.7177944183349609, 0.7762725353240967], "gt_se_raw": [239.0, 243.0], "gt_se_norm": [0.7470334768295288, 0.7762725353240967], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006227.pt"}
{"idx": 6229, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVMAwGXe3BY.mp4", "duration": 296.1959, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the pilot with the mustache contacts the tower to confirm the captain roster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 60.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the pilot with the mustache contacts the tower to confirm the captain roster.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 60.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [60.0, 105.0], "base_se_norm": [-0.594862699508667, -0.29100972414016724], "gt_se_raw": [60.0, 67.0], "gt_se_norm": [-0.594862699508667, -0.5475966930389404], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006229.pt"}
{"idx": 6231, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RVMAwGXe3BY.mp4", "duration": 296.1959, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when passengers are brought into the cockpit to meet the pilots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 212.0 - 236.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when passengers are brought into the cockpit to meet the pilots.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 212.0 - 236.0 seconds.<|im_end|>\n", "base_se_raw": [212.0, 236.0], "base_se_norm": [0.43148505687713623, 0.5935399532318115], "gt_se_raw": [212.0, 219.0], "gt_se_norm": [0.43148505687713623, 0.4787510633468628], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006231.pt"}
{"idx": 6233, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vVdW9CGI3Nc.mp4", "duration": 221.45456666666666, "prompt_text": "Please find the visual event described by the sentence 'When does Ernie show the towel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Ernie show the towel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 23.0], "base_se_norm": [-0.8284072279930115, -0.7922824621200562], "gt_se_raw": [19.0, 22.0], "gt_se_norm": [-0.8284072279930115, -0.8013136386871338], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006233.pt"}
{"idx": 6235, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vVdW9CGI3Nc.mp4", "duration": 221.45456666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where Ernie holds up a flashlight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where Ernie holds up a flashlight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 54.0], "base_se_norm": [-0.5394089221954346, -0.5123152732849121], "gt_se_raw": [51.0, 54.0], "gt_se_norm": [-0.5394089221954346, -0.5123152732849121], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006235.pt"}
{"idx": 6237, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vVdW9CGI3Nc.mp4", "duration": 221.45456666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where Ernie is in the bathtub under a red umbrella.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where Ernie is in the bathtub under a red umbrella.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 185.0], "base_se_norm": [0.47208523750305176, 0.670771598815918], "gt_se_raw": [164.0, 171.0], "gt_se_norm": [0.48111653327941895, 0.544334888458252], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006237.pt"}
{"idx": 6239, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZiz1i8EpSk.mp4", "duration": 212.6, "prompt_text": "Please find the visual event described by the sentence 'What is the time range for the interview with the man in the orange shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range for the interview with the man in the orange shirt?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 48.0], "base_se_norm": [-0.6237064599990845, -0.5484477877616882], "gt_se_raw": [40.0, 48.0], "gt_se_norm": [-0.6237064599990845, -0.5484477877616882], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006239.pt"}
{"idx": 6241, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GZiz1i8EpSk.mp4", "duration": 212.6, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a group of rescue personnel is walking down the middle of a street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a group of rescue personnel is walking down the middle of a street.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 85.0], "base_se_norm": [-0.24741297960281372, -0.20037633180618286], "gt_se_raw": [80.0, 85.0], "gt_se_norm": [-0.24741297960281372, -0.20037633180618286], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006241.pt"}
{"idx": 6243, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RnckZWnihFQ.mp4", "duration": 303.16953333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the video show a group of people playing a game with a blindfolded woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show a group of people playing a game with a blindfolded woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9802091121673584], "gt_se_raw": [0.0, 3.0], "gt_se_norm": [-1.0, -0.9802091121673584], "start_token_span": [73, 76], "end_token_span": [78, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006243.pt"}
{"idx": 6245, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RnckZWnihFQ.mp4", "duration": 303.16953333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the text 'Is Dependability An All or Nothing Quality?' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the text 'Is Dependability An All or Nothing Quality?' is on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 52.0], "base_se_norm": [-0.6833454370498657, -0.6569576263427734], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.6833454370498657, -0.6569576263427734], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006245.pt"}
{"idx": 6247, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RnckZWnihFQ.mp4", "duration": 303.16953333333333, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment where the text 'Repair Your Reputation By Changing Your Actions.' appears on an orange background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 247.0 - 251.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment where the text 'Repair Your Reputation By Changing Your Actions.' appears on an orange background.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 247.0 - 251.0 seconds.<|im_end|>\n", "base_se_raw": [247.0, 251.0], "base_se_norm": [0.6294513940811157, 0.655839204788208], "gt_se_raw": [247.0, 252.0], "gt_se_norm": [0.6294513940811157, 0.6624362468719482], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006247.pt"}
{"idx": 6249, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/snSwz_iPbQA.mp4", "duration": 228.10287499999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does a man hit a piece of meat with a hammer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a man hit a piece of meat with a hammer?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 94.0], "base_se_norm": [-0.21088236570358276, -0.17581045627593994], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.32486605644226074, -0.2897941470146179], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006249.pt"}
{"idx": 6251, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/snSwz_iPbQA.mp4", "duration": 228.10287499999998, "prompt_text": "Please find the visual event described by the sentence 'When does the man use a pair of sandals to create sounds in a box of gravel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man use a pair of sandals to create sounds in a box of gravel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 59.0], "base_se_norm": [-0.5177614688873291, -0.4826895594596863], "gt_se_raw": [55.0, 60.0], "gt_se_norm": [-0.5177614688873291, -0.47392159700393677], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006251.pt"}
{"idx": 6253, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3dHp0FdWMI0.mp4", "duration": 262.9, "prompt_text": "Please find the visual event described by the sentence 'When does the teacher give the girl a green ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the teacher give the girl a green ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 36.0], "base_se_norm": [-0.7945986986160278, -0.7261315584182739], "gt_se_raw": [27.0, 29.0], "gt_se_norm": [-0.7945986986160278, -0.7793837785720825], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006253.pt"}
{"idx": 6255, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3dHp0FdWMI0.mp4", "duration": 262.9, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the book page is turned to reveal the letter 'C'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 110.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the book page is turned to reveal the letter 'C'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 110.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 110.0], "base_se_norm": [-0.23164701461791992, -0.16317987442016602], "gt_se_raw": [110.0, 115.0], "gt_se_norm": [-0.16317987442016602, -0.1251426339149475], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006255.pt"}
{"idx": 6257, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3dHp0FdWMI0.mp4", "duration": 262.9, "prompt_text": "Please find the visual event described by the sentence 'At what point does the teacher start clapping to praise the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 243.0 - 258.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the teacher start clapping to praise the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 243.0 - 258.0 seconds.<|im_end|>\n", "base_se_raw": [243.0, 258.0], "base_se_norm": [0.8486117124557495, 0.9627234935760498], "gt_se_raw": [244.0, 248.0], "gt_se_norm": [0.8562191724777222, 0.8866490125656128], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006257.pt"}
{"idx": 6259, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8O5vcM_00t0.mp4", "duration": 231.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment when the dish soap is being added to the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment when the dish soap is being added to the water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 53.0], "base_se_norm": [-0.7061789035797119, -0.5419847369194031], "gt_se_raw": [37.0, 58.0], "gt_se_norm": [-0.6802535057067871, -0.4987757205963135], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006259.pt"}
{"idx": 6261, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8O5vcM_00t0.mp4", "duration": 231.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see multiple bubbles being blown inside each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see multiple bubbles being blown inside each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 128.0], "base_se_norm": [-0.22223824262619019, 0.10615003108978271], "gt_se_raw": [99.0, 128.0], "gt_se_norm": [-0.14446204900741577, 0.10615003108978271], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006261.pt"}
{"idx": 6263, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RghziC5GME.mp4", "duration": 204.1622888888889, "prompt_text": "Please find the visual event described by the sentence 'When can we see the girl talking to the boy while he is at the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the girl talking to the boy while he is at the counter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 70.0], "base_se_norm": [-0.9020386934280396, -0.3142710328102112], "gt_se_raw": [10.0, 20.0], "gt_se_norm": [-0.9020386934280396, -0.8040774464607239], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006263.pt"}
{"idx": 6265, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RghziC5GME.mp4", "duration": 204.1622888888889, "prompt_text": "Please find the visual event described by the sentence 'When does the boy fall to the ground after trying to kick the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 106.0 - 108.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy fall to the ground after trying to kick the ball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 106.0 - 108.0 seconds.<|im_end|>\n", "base_se_raw": [106.0, 108.0], "base_se_norm": [0.03838956356048584, 0.05798184871673584], "gt_se_raw": [106.0, 109.0], "gt_se_norm": [0.03838956356048584, 0.06777799129486084], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006265.pt"}
{"idx": 6267, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-RghziC5GME.mp4", "duration": 204.1622888888889, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the boy walks away with his head down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 84.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the boy walks away with his head down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 84.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [84.0, 102.0], "base_se_norm": [-0.1771252155303955, -0.0007949471473693848], "gt_se_raw": [175.0, 179.0], "gt_se_norm": [0.7143224477767944, 0.7535068988800049], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006267.pt"}
{"idx": 6269, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zZNygI-T1_c.mp4", "duration": 303.5949583333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the woman is on the phone at night, leaving a message.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the woman is on the phone at night, leaving a message.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 50.0], "base_se_norm": [-0.7562541961669922, -0.6706137657165527], "gt_se_raw": [38.0, 49.0], "gt_se_norm": [-0.7496664524078369, -0.677201509475708], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006269.pt"}
{"idx": 6271, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zZNygI-T1_c.mp4", "duration": 303.5949583333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do the man and woman start holding hands in the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 126.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the man and woman start holding hands in the park?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 126.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 126.0], "base_se_norm": [-0.19629764556884766, -0.16994673013687134], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [-0.19629764556884766, -0.1765344738960266], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006271.pt"}
{"idx": 6273, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f8PXvqYpGCM.mp4", "duration": 290.2232666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the boss, Brenda, first appear and yell at her staff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boss, Brenda, first appear and yell at her staff?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 31.0], "base_se_norm": [-0.8759575486183167, -0.7863713502883911], "gt_se_raw": [18.0, 31.0], "gt_se_norm": [-0.8759575486183167, -0.7863713502883911], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006273.pt"}
{"idx": 6275, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f8PXvqYpGCM.mp4", "duration": 290.2232666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video does it show that five years have gone by?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 189.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video does it show that five years have gone by?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 189.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [189.0, 193.0], "base_se_norm": [0.30244553089141846, 0.3300105333328247], "gt_se_raw": [189.0, 193.0], "gt_se_norm": [0.30244553089141846, 0.3300105333328247], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006275.pt"}
{"idx": 6277, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/f8PXvqYpGCM.mp4", "duration": 290.2232666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Donald tell Brenda that her nails are \"on fleek\"?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 92.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Donald tell Brenda that her nails are \"on fleek\"?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 92.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [92.0, 95.0], "base_se_norm": [-0.3660053610801697, -0.345331609249115], "gt_se_raw": [117.0, 121.0], "gt_se_norm": [-0.19372421503067017, -0.16615921258926392], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006277.pt"}
{"idx": 6279, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Lk8pXaDkh8.mp4", "duration": 249.2, "prompt_text": "Please find the visual event described by the sentence 'When does the character in the pink suit get hit by a bomb explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the character in the pink suit get hit by a bomb explosion?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 83.0], "base_se_norm": [-0.3739967942237854, -0.3338683843612671], "gt_se_raw": [78.0, 79.0], "gt_se_norm": [-0.3739967942237854, -0.36597108840942383], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006279.pt"}
{"idx": 6281, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8Lk8pXaDkh8.mp4", "duration": 249.2, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the pink boxer defeats the woman in the red dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 49.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the pink boxer defeats the woman in the red dress.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 49.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [49.0, 53.0], "base_se_norm": [-0.6067415475845337, -0.574638843536377], "gt_se_raw": [52.0, 54.0], "gt_se_norm": [-0.5826644897460938, -0.5666131973266602], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006281.pt"}
{"idx": 6283, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oFV9ayoss_o.mp4", "duration": 204.3043043043043, "prompt_text": "Please find the visual event described by the sentence 'When are three people seen doing yoga while using chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three people seen doing yoga while using chairs?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 31.0], "base_se_norm": [-0.8042136430740356, -0.6965311169624329], "gt_se_raw": [20.0, 30.0], "gt_se_norm": [-0.8042136430740356, -0.7063204050064087], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006283.pt"}
{"idx": 6285, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oFV9ayoss_o.mp4", "duration": 204.3043043043043, "prompt_text": "Please find the visual event described by the sentence 'At what point do a man and a woman perform a partner yoga pose while seated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 157.0 - 179.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do a man and a woman perform a partner yoga pose while seated?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 157.0 - 179.0 seconds.<|im_end|>\n", "base_se_raw": [157.0, 179.0], "base_se_norm": [0.5369230508804321, 0.7522881031036377], "gt_se_raw": [111.0, 119.0], "gt_se_norm": [0.08661437034606934, 0.16492891311645508], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006285.pt"}
{"idx": 6287, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oFV9ayoss_o.mp4", "duration": 204.3043043043043, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where a man and a woman, both seated, are doing a partner yoga pose, touching palms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 157.0 - 167.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where a man and a woman, both seated, are doing a partner yoga pose, touching palms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 157.0 - 167.0 seconds.<|im_end|>\n", "base_se_raw": [157.0, 167.0], "base_se_norm": [0.5369230508804321, 0.6348162889480591], "gt_se_raw": [156.0, 167.0], "gt_se_norm": [0.5271337032318115, 0.6348162889480591], "start_token_span": [77, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006287.pt"}
{"idx": 6289, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j7rHbHiHyo8.mp4", "duration": 273.73179166666665, "prompt_text": "Please find the visual event described by the sentence 'When is the little pikachu captured in a pokeball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 99.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the little pikachu captured in a pokeball?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 99.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [99.0, 106.0], "base_se_norm": [-0.27666419744491577, -0.22551923990249634], "gt_se_raw": [99.0, 106.0], "gt_se_norm": [-0.27666419744491577, -0.22551923990249634], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006289.pt"}
{"idx": 6291, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/j7rHbHiHyo8.mp4", "duration": 273.73179166666665, "prompt_text": "Please find the visual event described by the sentence 'Show me when the pikachu is in the hospital room being healed by the nurse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 148.0 - 181.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the pikachu is in the hospital room being healed by the nurse.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 148.0 - 181.0 seconds.<|im_end|>\n", "base_se_raw": [148.0, 181.0], "base_se_norm": [0.08135056495666504, 0.32246243953704834], "gt_se_raw": [147.0, 152.0], "gt_se_norm": [0.0740441083908081, 0.11057615280151367], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006291.pt"}
{"idx": 6293, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6M_WjPuZIeA.mp4", "duration": 258.8753222222222, "prompt_text": "Please find the visual event described by the sentence 'When does the clown emerge from the sewer drain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the clown emerge from the sewer drain?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 24.0], "base_se_norm": [-0.8377597332000732, -0.8145825862884521], "gt_se_raw": [21.0, 23.0], "gt_se_norm": [-0.8377597332000732, -0.8223083019256592], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006293.pt"}
{"idx": 6295, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6M_WjPuZIeA.mp4", "duration": 258.8753222222222, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the child in the yellow raincoat bumps into the construction barrier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 14.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the child in the yellow raincoat bumps into the construction barrier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 14.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 14.0], "base_se_norm": [-0.9150170087814331, -0.891839861869812], "gt_se_raw": [12.0, 13.0], "gt_se_norm": [-0.9072912931442261, -0.899565577507019], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006295.pt"}
{"idx": 6297, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6M_WjPuZIeA.mp4", "duration": 258.8753222222222, "prompt_text": "Please find the visual event described by the sentence 'When does the child sit in the big yellow boat and float away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 232.0 - 237.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the child sit in the big yellow boat and float away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 232.0 - 237.0 seconds.<|im_end|>\n", "base_se_raw": [232.0, 237.0], "base_se_norm": [0.7923685312271118, 0.830997109413147], "gt_se_raw": [232.0, 237.0], "gt_se_norm": [0.7923685312271118, 0.830997109413147], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006297.pt"}
{"idx": 6299, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/38y_1EWIE9I.mp4", "duration": 281.4, "prompt_text": "Please find the visual event described by the sentence 'At what point does the old lady finally get the cookie out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the old lady finally get the cookie out of the machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 71.0], "base_se_norm": [-0.5167021751403809, -0.4953802227973938], "gt_se_raw": [68.0, 71.0], "gt_se_norm": [-0.5167021751403809, -0.4953802227973938], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006299.pt"}
{"idx": 6301, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/38y_1EWIE9I.mp4", "duration": 281.4, "prompt_text": "Please find the visual event described by the sentence 'When does the young man break the cookie in half and share it with the old lady?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the young man break the cookie in half and share it with the old lady?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 160.0], "base_se_norm": [0.08031272888183594, 0.13717126846313477], "gt_se_raw": [153.0, 157.0], "gt_se_norm": [0.08742010593414307, 0.11584937572479248], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006301.pt"}
{"idx": 6303, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pt97zl_gwnk.mp4", "duration": 290.8739166666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the woman holds up the boy's hands to show they are red.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the woman holds up the boy's hands to show they are red.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 41.0], "base_se_norm": [-0.7937250733375549, -0.7180908918380737], "gt_se_raw": [29.0, 33.0], "gt_se_norm": [-0.8006008863449097, -0.773097574710846], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006303.pt"}
{"idx": 6305, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pt97zl_gwnk.mp4", "duration": 290.8739166666667, "prompt_text": "Please find the visual event described by the sentence 'When is the boy stirring the mixture in the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy stirring the mixture in the measuring cup?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 65.0], "base_se_norm": [-0.6012017726898193, -0.5530710220336914], "gt_se_raw": [80.0, 88.0], "gt_se_norm": [-0.4499335289001465, -0.39492684602737427], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006305.pt"}
{"idx": 6307, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Pt97zl_gwnk.mp4", "duration": 290.8739166666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman put her face in the pink slime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 271.0 - 276.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman put her face in the pink slime?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 271.0 - 276.0 seconds.<|im_end|>\n", "base_se_raw": [271.0, 276.0], "base_se_norm": [0.8633501529693604, 0.8977293968200684], "gt_se_raw": [272.0, 276.0], "gt_se_norm": [0.8702260255813599, 0.8977293968200684], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006307.pt"}
{"idx": 6309, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/amsqeYOk--w.mp4", "duration": 314.75, "prompt_text": "Please find the visual event described by the sentence 'A little girl in a pink t-shirt eats the marshmallow. What is her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A little girl in a pink t-shirt eats the marshmallow. What is her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 76.0], "base_se_norm": [-0.5552025437355042, -0.5170770883560181], "gt_se_raw": [70.0, 74.0], "gt_se_norm": [-0.5552025437355042, -0.5297855138778687], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006309.pt"}
{"idx": 6311, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/amsqeYOk--w.mp4", "duration": 314.75, "prompt_text": "Please find the visual event described by the sentence 'When does the boy in the dark blue sweatshirt decide to eat the marshmallow right away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 170.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy in the dark blue sweatshirt decide to eat the marshmallow right away?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 170.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 170.0], "base_se_norm": [0.05480539798736572, 0.08022236824035645], "gt_se_raw": [153.0, 155.0], "gt_se_norm": [-0.02779984474182129, -0.015091359615325928], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006311.pt"}
{"idx": 6313, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFu4wn-s-eg.mp4", "duration": 207.47400000000002, "prompt_text": "Please find the visual event described by the sentence 'When are the two boys walking along the country road at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two boys walking along the country road at the beginning of the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 12.0], "base_se_norm": [-1.0, -0.8843228816986084], "gt_se_raw": [0.0, 12.0], "gt_se_norm": [-1.0, -0.8843228816986084], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006313.pt"}
{"idx": 6315, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFu4wn-s-eg.mp4", "duration": 207.47400000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man discovers the first silver dollar in his shoe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 103.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man discovers the first silver dollar in his shoe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 103.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 103.0], "base_se_norm": [-0.03602379560470581, -0.00710451602935791], "gt_se_raw": [99.0, 106.0], "gt_se_norm": [-0.045663535594940186, 0.021814823150634766], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006315.pt"}
{"idx": 6317, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFu4wn-s-eg.mp4", "duration": 207.47400000000002, "prompt_text": "Please find the visual event described by the sentence 'Show the shot where the boys are hiding and the younger boy is visibly moved.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 171.0 - 181.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the shot where the boys are hiding and the younger boy is visibly moved.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 171.0 - 181.0 seconds.<|im_end|>\n", "base_se_raw": [171.0, 181.0], "base_se_norm": [0.6483993530273438, 0.7447969913482666], "gt_se_raw": [173.0, 180.0], "gt_se_norm": [0.6676788330078125, 0.7351571321487427], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006317.pt"}
{"idx": 6319, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gAaWm6-4dro.mp4", "duration": 325.1915333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where two women are conversing outdoors in front of the CNIB building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where two women are conversing outdoors in front of the CNIB building.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 40.0], "base_se_norm": [-0.8400942087173462, -0.7539911270141602], "gt_se_raw": [26.0, 40.0], "gt_se_norm": [-0.8400942087173462, -0.7539911270141602], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006319.pt"}
{"idx": 6321, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/gAaWm6-4dro.mp4", "duration": 325.1915333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the time range when the three women are exiting the building through the sliding glass doors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 106.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time range when the three women are exiting the building through the sliding glass doors.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 106.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 106.0], "base_se_norm": [-0.37882763147354126, -0.3480765223503113], "gt_se_raw": [101.0, 105.0], "gt_se_norm": [-0.37882763147354126, -0.3542267084121704], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006321.pt"}
{"idx": 6323, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/itHff9ErfpQ.mp4", "duration": 214.84796666666668, "prompt_text": "Please find the visual event described by the sentence 'When is the glue being poured into the white bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 39.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the glue being poured into the white bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 39.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [39.0, 50.0], "base_se_norm": [-0.6369525790214539, -0.5345546007156372], "gt_se_raw": [38.0, 51.0], "gt_se_norm": [-0.64626145362854, -0.5252456665039062], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006323.pt"}
{"idx": 6325, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/itHff9ErfpQ.mp4", "duration": 214.84796666666668, "prompt_text": "Please find the visual event described by the sentence 'At what moment is the green food coloring dropped into the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 78.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is the green food coloring dropped into the mixture?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 78.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [78.0, 82.0], "base_se_norm": [-0.2739051580429077, -0.23666948080062866], "gt_se_raw": [78.0, 82.0], "gt_se_norm": [-0.2739051580429077, -0.23666948080062866], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006325.pt"}
{"idx": 6327, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/itHff9ErfpQ.mp4", "duration": 214.84796666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the boy pick up the finished slime with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy pick up the finished slime with his hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 168.0], "base_se_norm": [0.4987342357635498, 0.563896656036377], "gt_se_raw": [162.0, 165.0], "gt_se_norm": [0.5080431699752808, 0.5359698534011841], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006327.pt"}
{"idx": 6329, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ndA4YL-bBAQ.mp4", "duration": 328.88, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the fly lands on Betty's nose, causing her to sneeze.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the fly lands on Betty's nose, causing her to sneeze.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 72.0], "base_se_norm": [-0.5803940296173096, -0.5621503591537476], "gt_se_raw": [69.0, 72.0], "gt_se_norm": [-0.5803940296173096, -0.5621503591537476], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006329.pt"}
{"idx": 6331, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ndA4YL-bBAQ.mp4", "duration": 328.88, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when Betty throws dough at the portrait on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 170.0 - 175.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when Betty throws dough at the portrait on the wall.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 170.0 - 175.0 seconds.<|im_end|>\n", "base_se_raw": [170.0, 175.0], "base_se_norm": [0.03381168842315674, 0.06421792507171631], "gt_se_raw": [171.0, 176.0], "gt_se_norm": [0.03989291191101074, 0.07029914855957031], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006331.pt"}
{"idx": 6333, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOxOgGkAjS8.mp4", "duration": 275.6, "prompt_text": "Please find the visual event described by the sentence 'When are the cauliflower florets soaked in hot water to prepare them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the cauliflower florets soaked in hot water to prepare them?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 37.0], "base_se_norm": [-0.8621190190315247, -0.7314949035644531], "gt_se_raw": [20.0, 35.0], "gt_se_norm": [-0.8548620939254761, -0.7460087537765503], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006333.pt"}
{"idx": 6335, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOxOgGkAjS8.mp4", "duration": 275.6, "prompt_text": "Please find the visual event described by the sentence 'At what point are the battered cauliflower pieces deep-fried until golden and crispy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the battered cauliflower pieces deep-fried until golden and crispy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 134.0], "base_se_norm": [-0.2743105888366699, -0.027576208114624023], "gt_se_raw": [99.0, 134.0], "gt_se_norm": [-0.2815675139427185, -0.027576208114624023], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006335.pt"}
{"idx": 6337, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NOxOgGkAjS8.mp4", "duration": 275.6, "prompt_text": "Please find the visual event described by the sentence 'When is the final step where spring onions are added as a garnish and mixed in?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 244.0 - 256.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the final step where spring onions are added as a garnish and mixed in?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 244.0 - 256.0 seconds.<|im_end|>\n", "base_se_raw": [244.0, 256.0], "base_se_norm": [0.7706820964813232, 0.8577648401260376], "gt_se_raw": [244.0, 256.0], "gt_se_norm": [0.7706820964813232, 0.8577648401260376], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006337.pt"}
{"idx": 6339, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RpJd0HeaPHo.mp4", "duration": 247.31373333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the woman walking with her guide dog in a snowy outdoor scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman walking with her guide dog in a snowy outdoor scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 82.0], "base_se_norm": [-0.3773091435432434, -0.33687466382980347], "gt_se_raw": [58.0, 59.0], "gt_se_norm": [-0.5309601426124573, -0.5228732824325562], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006339.pt"}
{"idx": 6341, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RpJd0HeaPHo.mp4", "duration": 247.31373333333335, "prompt_text": "Please find the visual event described by the sentence 'At what time is the speaker lying in bed and petting her dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 37.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is the speaker lying in bed and petting her dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 37.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [37.0, 50.0], "base_se_norm": [-0.7007849216461182, -0.5956553220748901], "gt_se_raw": [38.0, 41.0], "gt_se_norm": [-0.6926980018615723, -0.6684373617172241], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006341.pt"}
{"idx": 6343, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jX9YVuZ8hZc.mp4", "duration": 265.09816666666666, "prompt_text": "Please find the visual event described by the sentence 'When is the man programming at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man programming at his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 42.0], "base_se_norm": [-0.8943787813186646, -0.6831362247467041], "gt_se_raw": [25.0, 35.0], "gt_se_norm": [-0.8113906383514404, -0.7359468936920166], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006343.pt"}
{"idx": 6345, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jX9YVuZ8hZc.mp4", "duration": 265.09816666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the man makes himself a cup of coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 101.0 - 107.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the man makes himself a cup of coffee.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 101.0 - 107.0 seconds.<|im_end|>\n", "base_se_raw": [101.0, 107.0], "base_se_norm": [-0.23801815509796143, -0.19275188446044922], "gt_se_raw": [101.0, 107.0], "gt_se_norm": [-0.23801815509796143, -0.19275188446044922], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006345.pt"}
{"idx": 6347, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jX9YVuZ8hZc.mp4", "duration": 265.09816666666666, "prompt_text": "Please find the visual event described by the sentence 'At what moment is the man chased by the lights and falls down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is the man chased by the lights and falls down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 153.0], "base_se_norm": [0.09393429756164551, 0.1542893648147583], "gt_se_raw": [144.0, 150.0], "gt_se_norm": [0.08639001846313477, 0.13165616989135742], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006347.pt"}
{"idx": 6349, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8toX8ZjCDVA.mp4", "duration": 298.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'When is cream, sugar, and condensed milk added to the pot on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is cream, sugar, and condensed milk added to the pot on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 76.0], "base_se_norm": [-0.5842646360397339, -0.4903889298439026], "gt_se_raw": [62.0, 75.0], "gt_se_norm": [-0.5842646360397339, -0.4970943331718445], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006349.pt"}
{"idx": 6351, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8toX8ZjCDVA.mp4", "duration": 298.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'At what point is the chocolate log being dusted with cocoa powder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 215.0 - 223.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the chocolate log being dusted with cocoa powder?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 215.0 - 223.0 seconds.<|im_end|>\n", "base_se_raw": [215.0, 223.0], "base_se_norm": [0.44166290760040283, 0.49530625343322754], "gt_se_raw": [215.0, 228.0], "gt_se_norm": [0.44166290760040283, 0.528833270072937], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006351.pt"}
{"idx": 6353, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yK17-IATTE8.mp4", "duration": 204.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the villain in red armor grab the man in blue and throw him down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the villain in red armor grab the man in blue and throw him down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.7456797957420349, -0.7065536379814148], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.7358982563018799, -0.7065536379814148], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006353.pt"}
{"idx": 6355, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yK17-IATTE8.mp4", "duration": 204.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman touch the stone with glowing patterns on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman touch the stone with glowing patterns on it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 75.0], "base_se_norm": [-0.28594714403152466, -0.2663840651512146], "gt_se_raw": [73.0, 75.0], "gt_se_norm": [-0.28594714403152466, -0.2663840651512146], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006355.pt"}
{"idx": 6357, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yK17-IATTE8.mp4", "duration": 204.46666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the king knights the man and the woman with his sword.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 193.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the king knights the man and the woman with his sword.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 193.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [193.0, 201.0], "base_se_norm": [0.8878383636474609, 0.9660906791687012], "gt_se_raw": [193.0, 201.0], "gt_se_norm": [0.8878383636474609, 0.9660906791687012], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006357.pt"}
{"idx": 6359, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yd_1z6OsrE.mp4", "duration": 276.23429166666665, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the NNELS logo with red hexagons is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 23.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the NNELS logo with red hexagons is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 23.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [23.0, 28.0], "base_se_norm": [-0.8334746956825256, -0.7972735166549683], "gt_se_raw": [23.0, 28.0], "gt_se_norm": [-0.8334746956825256, -0.7972735166549683], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006359.pt"}
{"idx": 6361, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3yd_1z6OsrE.mp4", "duration": 276.23429166666665, "prompt_text": "Please find the visual event described by the sentence 'Locate the animation where a hand clicks on the number 1, revealing the text 'Charles Darwin, 1859'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 71.0 - 75.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the animation where a hand clicks on the number 1, revealing the text 'Charles Darwin, 1859'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 71.0 - 75.0 seconds.<|im_end|>\n", "base_se_raw": [71.0, 75.0], "base_se_norm": [-0.48594361543655396, -0.45698267221450806], "gt_se_raw": [71.0, 75.0], "gt_se_norm": [-0.48594361543655396, -0.45698267221450806], "start_token_span": [82, 86], "end_token_span": [88, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006361.pt"}
{"idx": 6363, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6US4AyvEO_A.mp4", "duration": 262.3037083333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two men having a conversation next to the mailboxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men having a conversation next to the mailboxes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 50.0], "base_se_norm": [-0.9008782505989075, -0.6187625527381897], "gt_se_raw": [14.0, 49.0], "gt_se_norm": [-0.89325350522995, -0.6263872981071472], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006363.pt"}
{"idx": 6365, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6US4AyvEO_A.mp4", "duration": 262.3037083333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the blue shirt get angry after seeing his dead cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 133.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the blue shirt get angry after seeing his dead cat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 133.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [133.0, 158.0], "base_se_norm": [0.0140916109085083, 0.20471036434173584], "gt_se_raw": [133.0, 143.0], "gt_se_norm": [0.0140916109085083, 0.09033918380737305], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006365.pt"}
{"idx": 6367, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6US4AyvEO_A.mp4", "duration": 262.3037083333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the grey shirt defecate on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 238.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the grey shirt defecate on the lawn?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 238.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 238.0], "base_se_norm": [0.7003190517425537, 0.814690351486206], "gt_se_raw": [231.0, 246.0], "gt_se_norm": [0.7613170146942139, 0.8756883144378662], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006367.pt"}
{"idx": 6369, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pXmHP-7FcP0.mp4", "duration": 210.5, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man in the patterned shirt first speaks to clarify his role.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man in the patterned shirt first speaks to clarify his role.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 46.0], "base_se_norm": [-0.6674584150314331, -0.5629453659057617], "gt_se_raw": [35.0, 38.0], "gt_se_norm": [-0.6674584150314331, -0.6389548778533936], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006369.pt"}
{"idx": 6371, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pXmHP-7FcP0.mp4", "duration": 210.5, "prompt_text": "Please find the visual event described by the sentence 'When does the woman roll her eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 63.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman roll her eyes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 63.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 63.0], "base_se_norm": [-0.4394299387931824, -0.40142518281936646], "gt_se_raw": [97.0, 98.0], "gt_se_norm": [-0.07838481664657593, -0.06888359785079956], "start_token_span": [62, 66], "end_token_span": [68, 72], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006371.pt"}
{"idx": 6373, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TCw-iEJoVfY.mp4", "duration": 236.6530833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the detective aiming his gun at multiple reflections of Carlo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 31.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the detective aiming his gun at multiple reflections of Carlo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 31.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [31.0, 48.0], "base_se_norm": [-0.7380131483078003, -0.5943429470062256], "gt_se_raw": [85.0, 90.0], "gt_se_norm": [-0.2816488742828369, -0.2393929362297058], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006373.pt"}
{"idx": 6375, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TCw-iEJoVfY.mp4", "duration": 236.6530833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the instance where the detective points his gun at the head of a man with blonde hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 132.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the instance where the detective points his gun at the head of a man with blonde hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 132.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [132.0, 164.0], "base_se_norm": [0.11555707454681396, 0.3859950304031372], "gt_se_raw": [105.0, 110.0], "gt_se_norm": [-0.1126251220703125, -0.07036912441253662], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006375.pt"}
{"idx": 6377, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TCw-iEJoVfY.mp4", "duration": 236.6530833333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the detective grabs a man's lapel and yells at him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 187.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the detective grabs a man's lapel and yells at him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 187.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [187.0, 198.0], "base_se_norm": [0.5803724527359009, 0.6733355522155762], "gt_se_raw": [180.0, 185.0], "gt_se_norm": [0.5212141275405884, 0.5634700059890747], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006377.pt"}
{"idx": 6379, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HHsmRoHE7K0.mp4", "duration": 230.23, "prompt_text": "Please find the visual event described by the sentence 'When are the empanadas being filled and folded by a machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 169.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the empanadas being filled and folded by a machine?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 169.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 169.0], "base_se_norm": [0.20748817920684814, 0.46809709072113037], "gt_se_raw": [144.0, 169.0], "gt_se_norm": [0.2509230375289917, 0.46809709072113037], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006379.pt"}
{"idx": 6381, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/HHsmRoHE7K0.mp4", "duration": 230.23, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a person adds ground beef to a pot of onions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a person adds ground beef to a pot of onions.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 50.0], "base_se_norm": [-0.5917125940322876, -0.5656517744064331], "gt_se_raw": [48.0, 50.0], "gt_se_norm": [-0.5830256938934326, -0.5656517744064331], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006381.pt"}
{"idx": 6383, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nkpKb2y5BLQ.mp4", "duration": 263.36310000000003, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man with the gray beard and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 120.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man with the gray beard and glasses speaking?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 120.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 120.0], "base_se_norm": [-0.3165329694747925, -0.08871060609817505], "gt_se_raw": [90.0, 121.0], "gt_se_norm": [-0.3165329694747925, -0.08111649751663208], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006383.pt"}
{"idx": 6385, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nkpKb2y5BLQ.mp4", "duration": 263.36310000000003, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the woman wearing a pearl necklace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the woman wearing a pearl necklace?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 146.0], "base_se_norm": [-0.08111649751663208, 0.10873544216156006], "gt_se_raw": [121.0, 146.0], "gt_se_norm": [-0.08111649751663208, 0.10873544216156006], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006385.pt"}
{"idx": 6387, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/nkpKb2y5BLQ.mp4", "duration": 263.36310000000003, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when Tom Irish, the Assistant Professor of English, is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 14.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when Tom Irish, the Assistant Professor of English, is on screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 14.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [14.0, 48.0], "base_se_norm": [-0.8936828970909119, -0.6354842185974121], "gt_se_raw": [14.0, 48.0], "gt_se_norm": [-0.8936828970909119, -0.6354842185974121], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006387.pt"}
{"idx": 6389, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BDnXdeoiYRE.mp4", "duration": 312.2285833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man begin to list names for the puppies?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 91.0 - 98.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man begin to list names for the puppies?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 91.0 - 98.0 seconds.<|im_end|>\n", "base_se_raw": [91.0, 98.0], "base_se_norm": [-0.4170936942100525, -0.37225478887557983], "gt_se_raw": [92.0, 99.0], "gt_se_norm": [-0.4106881618499756, -0.36584919691085815], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006389.pt"}
{"idx": 6391, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BDnXdeoiYRE.mp4", "duration": 312.2285833333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see a white puppy and a black puppy playing together near the man's shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 141.0 - 146.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a white puppy and a black puppy playing together near the man's shoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 141.0 - 146.0 seconds.<|im_end|>\n", "base_se_raw": [141.0, 146.0], "base_se_norm": [-0.09681552648544312, -0.06478774547576904], "gt_se_raw": [141.0, 146.0], "gt_se_norm": [-0.09681552648544312, -0.06478774547576904], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006391.pt"}
{"idx": 6393, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OIDEGN4Js40.mp4", "duration": 242.1585777777778, "prompt_text": "Please find the visual event described by the sentence 'At what point does Biden show Obama the sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does Biden show Obama the sunglasses?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 24.0], "base_se_norm": [-0.8595961332321167, -0.8017827868461609], "gt_se_raw": [19.0, 27.0], "gt_se_norm": [-0.8430780172348022, -0.7770056128501892], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006393.pt"}
{"idx": 6395, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OIDEGN4Js40.mp4", "duration": 242.1585777777778, "prompt_text": "Please find the visual event described by the sentence 'When is Obama having a conversation with the DMV clerk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Obama having a conversation with the DMV clerk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 85.0], "base_se_norm": [-0.5292341113090515, -0.29798072576522827], "gt_se_raw": [60.0, 68.0], "gt_se_norm": [-0.5044569969177246, -0.4383845329284668], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006395.pt"}
{"idx": 6397, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OIDEGN4Js40.mp4", "duration": 242.1585777777778, "prompt_text": "Please find the visual event described by the sentence 'When are the two men watching the movie in the theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men watching the movie in the theater?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 151.0], "base_se_norm": [0.13148987293243408, 0.24711668491363525], "gt_se_raw": [137.0, 151.0], "gt_se_norm": [0.13148987293243408, 0.24711668491363525], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006397.pt"}
{"idx": 6399, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3U-umc-pE20.mp4", "duration": 258.68, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the dog catches the tennis ball in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 81.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the dog catches the tennis ball in its mouth.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 81.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 81.0], "base_se_norm": [-0.41240137815475464, -0.37374359369277954], "gt_se_raw": [77.0, 79.0], "gt_se_norm": [-0.4046698212623596, -0.3892067074775696], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006399.pt"}
{"idx": 6401, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3U-umc-pE20.mp4", "duration": 258.68, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment the dog rests its head on the man's lap while they are sitting on the pier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment the dog rests its head on the man's lap while they are sitting on the pier.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 130.0], "base_se_norm": [-0.03355497121810913, 0.005102872848510742], "gt_se_raw": [125.0, 129.0], "gt_se_norm": [-0.03355497121810913, -0.0026286840438842773], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006401.pt"}
{"idx": 6403, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzRkHkssdME.mp4", "duration": 209.67613333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the clouds hitting the snowy mountains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the clouds hitting the snowy mountains?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 26.0], "base_se_norm": [-0.8473836779594421, -0.7519984841346741], "gt_se_raw": [16.0, 27.0], "gt_se_norm": [-0.8473836779594421, -0.7424599528312683], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006403.pt"}
{"idx": 6405, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzRkHkssdME.mp4", "duration": 209.67613333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shots of various jellyfish-like creatures in the dark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shots of various jellyfish-like creatures in the dark.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 134.0], "base_se_norm": [-0.44676584005355835, 0.2781616449356079], "gt_se_raw": [58.0, 73.0], "gt_se_norm": [-0.44676584005355835, -0.30368804931640625], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006405.pt"}
{"idx": 6407, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vzRkHkssdME.mp4", "duration": 209.67613333333333, "prompt_text": "Please find the visual event described by the sentence 'When does Will Smith appear and speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Will Smith appear and speak to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 147.0], "base_se_norm": [0.28770017623901367, 0.40216243267059326], "gt_se_raw": [135.0, 147.0], "gt_se_norm": [0.28770017623901367, 0.40216243267059326], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006407.pt"}
{"idx": 6409, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3XGUnjDRUQ.mp4", "duration": 267.267, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when a large number of discarded tires can be seen at the bottom of the ocean.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when a large number of discarded tires can be seen at the bottom of the ocean.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.8578200936317444, -0.8054380416870117], "gt_se_raw": [19.0, 25.0], "gt_se_norm": [-0.8578200936317444, -0.8129211664199829], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006409.pt"}
{"idx": 6411, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/l3XGUnjDRUQ.mp4", "duration": 267.267, "prompt_text": "Please find the visual event described by the sentence 'At what point in the video is a turtle seen with its body constricted by a plastic ring?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 64.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point in the video is a turtle seen with its body constricted by a plastic ring?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 64.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [64.0, 66.0], "base_se_norm": [-0.5210781693458557, -0.5061118602752686], "gt_se_raw": [64.0, 66.0], "gt_se_norm": [-0.5210781693458557, -0.5061118602752686], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006411.pt"}
{"idx": 6413, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pPdevJTGAYY.mp4", "duration": 272.12, "prompt_text": "Please find the visual event described by the sentence 'When can we see the man with the wild grey hair talking in his office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 9.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the man with the wild grey hair talking in his office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 9.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 9.0], "base_se_norm": [-1.0, -0.9338527321815491], "gt_se_raw": [0.0, 9.0], "gt_se_norm": [-1.0, -0.9338527321815491], "start_token_span": [70, 73], "end_token_span": [75, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006413.pt"}
{"idx": 6415, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pPdevJTGAYY.mp4", "duration": 272.12, "prompt_text": "Please find the visual event described by the sentence 'When does the scientist in the green coat appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the scientist in the green coat appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 26.0], "base_se_norm": [-0.8824048042297363, -0.8089078664779663], "gt_se_raw": [16.0, 22.0], "gt_se_norm": [-0.8824048042297363, -0.8383066058158875], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006415.pt"}
{"idx": 6417, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pPdevJTGAYY.mp4", "duration": 272.12, "prompt_text": "Please find the visual event described by the sentence 'When is the moment the man in the white lab coat wets the wall with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 248.0 - 257.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the moment the man in the white lab coat wets the wall with water?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 248.0 - 257.0 seconds.<|im_end|>\n", "base_se_raw": [248.0, 257.0], "base_se_norm": [0.8227252960205078, 0.8888726234436035], "gt_se_raw": [249.0, 258.0], "gt_se_norm": [0.8300750255584717, 0.8962222337722778], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006417.pt"}
{"idx": 6419, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/88tK5c0wgH4.mp4", "duration": 219.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does molten lava erupt from a volcano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 81.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does molten lava erupt from a volcano?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 81.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [81.0, 104.0], "base_se_norm": [-0.2611736059188843, -0.051383376121520996], "gt_se_raw": [97.0, 100.0], "gt_se_norm": [-0.11523258686065674, -0.08786863088607788], "start_token_span": [66, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006419.pt"}
{"idx": 6421, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/88tK5c0wgH4.mp4", "duration": 219.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'When is Albert Einstein's equation E=mc\u00b2 shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 146.0 - 150.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is Albert Einstein's equation E=mc\u00b2 shown on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 146.0 - 150.0 seconds.<|im_end|>\n", "base_se_raw": [146.0, 150.0], "base_se_norm": [0.3317117691040039, 0.36819708347320557], "gt_se_raw": [147.0, 150.0], "gt_se_norm": [0.34083306789398193, 0.36819708347320557], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006421.pt"}
{"idx": 6423, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B1yJuGQOUaY.mp4", "duration": 203.0, "prompt_text": "Please find the visual event described by the sentence 'When does the man make the first latte art in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man make the first latte art in the video?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 114.0], "base_se_norm": [-0.2216748595237732, 0.1231527328491211], "gt_se_raw": [81.0, 93.0], "gt_se_norm": [-0.20197045803070068, -0.0837438702583313], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006423.pt"}
{"idx": 6425, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B1yJuGQOUaY.mp4", "duration": 203.0, "prompt_text": "Please find the visual event described by the sentence 'Show the time the man prepares the coffee grounds and tamps them down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 2.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time the man prepares the coffee grounds and tamps them down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 2.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [2.0, 16.0], "base_se_norm": [-0.9802955389022827, -0.8423645496368408], "gt_se_raw": [21.0, 26.0], "gt_se_norm": [-0.7931034564971924, -0.743842363357544], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006425.pt"}
{"idx": 6427, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/B1yJuGQOUaY.mp4", "duration": 203.0, "prompt_text": "Please find the visual event described by the sentence 'Show the final shot where all the finished latte art cups are displayed in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 175.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the final shot where all the finished latte art cups are displayed in a row.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 175.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [175.0, 201.0], "base_se_norm": [0.7241379022598267, 0.9802955389022827], "gt_se_raw": [199.0, 202.0], "gt_se_norm": [0.9605910778045654, 0.9901478290557861], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006427.pt"}
{"idx": 6429, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OPdbdjctx2I.mp4", "duration": 270.4702, "prompt_text": "Please find the visual event described by the sentence 'Show the point where a park employee talks to Kevin Hart in front of a height requirement sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 90.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the point where a park employee talks to Kevin Hart in front of a height requirement sign.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 90.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [90.0, 104.0], "base_se_norm": [-0.3344923257827759, -0.23096895217895508], "gt_se_raw": [91.0, 100.0], "gt_se_norm": [-0.3270978331565857, -0.26054704189300537], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006429.pt"}
{"idx": 6431, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OPdbdjctx2I.mp4", "duration": 270.4702, "prompt_text": "Please find the visual event described by the sentence 'When does Jimmy Fallon complain about a bug hitting him in the face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 194.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Jimmy Fallon complain about a bug hitting him in the face?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 194.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 194.0], "base_se_norm": [0.3384099006652832, 0.43453872203826904], "gt_se_raw": [180.0, 185.0], "gt_se_norm": [0.33101534843444824, 0.3679879903793335], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006431.pt"}
{"idx": 6433, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y8v2hvtM9_c.mp4", "duration": 227.88, "prompt_text": "Please find the visual event described by the sentence 'When are the two men squatting and conversing by the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two men squatting and conversing by the river?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 147.0], "base_se_norm": [-0.9912234544754028, 0.2901526689529419], "gt_se_raw": [1.0, 20.0], "gt_se_norm": [-0.9912234544754028, -0.8244690299034119], "start_token_span": [68, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006433.pt"}
{"idx": 6435, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y8v2hvtM9_c.mp4", "duration": 227.88, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where one man kneels behind the other and places his hand on his neck while talking to him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 88.0 - 101.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where one man kneels behind the other and places his hand on his neck while talking to him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 88.0 - 101.0 seconds.<|im_end|>\n", "base_se_raw": [88.0, 101.0], "base_se_norm": [-0.22766369581222534, -0.11356854438781738], "gt_se_raw": [89.0, 100.0], "gt_se_norm": [-0.21888715028762817, -0.12234508991241455], "start_token_span": [78, 82], "end_token_span": [84, 89], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006435.pt"}
{"idx": 6437, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y8v2hvtM9_c.mp4", "duration": 227.88, "prompt_text": "Please find the visual event described by the sentence 'When can we see the two men working together in a field, lifting sacks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the two men working together in a field, lifting sacks?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 188.0], "base_se_norm": [0.6236615180969238, 0.6499911546707153], "gt_se_raw": [182.0, 189.0], "gt_se_norm": [0.5973318815231323, 0.6587677001953125], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006437.pt"}
{"idx": 6439, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_wsnc8IdCQ.mp4", "duration": 255.24, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the man is outside his house with his dogs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the man is outside his house with his dogs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 51.0], "base_se_norm": [-0.6787337064743042, -0.6003761291503906], "gt_se_raw": [41.0, 51.0], "gt_se_norm": [-0.6787337064743042, -0.6003761291503906], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006439.pt"}
{"idx": 6441, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/b_wsnc8IdCQ.mp4", "duration": 255.24, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the wheelchair use eye-tracking technology to make a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 156.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the wheelchair use eye-tracking technology to make a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 156.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 156.0], "base_se_norm": [0.18320012092590332, 0.22237896919250488], "gt_se_raw": [148.0, 156.0], "gt_se_norm": [0.15969276428222656, 0.22237896919250488], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006441.pt"}
{"idx": 6443, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5itl-NItMdI.mp4", "duration": 270.20353687020355, "prompt_text": "Please find the visual event described by the sentence 'When are the strawberries being mashed in a glass jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 46.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the strawberries being mashed in a glass jar?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 46.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [46.0, 68.0], "base_se_norm": [-0.659515917301178, -0.49667567014694214], "gt_se_raw": [45.0, 67.0], "gt_se_norm": [-0.6669177412986755, -0.5040775537490845], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006443.pt"}
{"idx": 6445, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5itl-NItMdI.mp4", "duration": 270.20353687020355, "prompt_text": "Please find the visual event described by the sentence 'At what point is the strawberry mixture being cooked on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the strawberry mixture being cooked on the stove?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 141.0], "base_se_norm": [-0.2820226550102234, 0.043657779693603516], "gt_se_raw": [94.0, 140.0], "gt_se_norm": [-0.30422818660736084, 0.03625595569610596], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006445.pt"}
{"idx": 6447, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5itl-NItMdI.mp4", "duration": 270.20353687020355, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the final strawberry milk is poured into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 251.0 - 265.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the final strawberry milk is poured into a glass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 251.0 - 265.0 seconds.<|im_end|>\n", "base_se_raw": [251.0, 265.0], "base_se_norm": [0.8578587770462036, 0.961484432220459], "gt_se_raw": [251.0, 265.0], "gt_se_norm": [0.8578587770462036, 0.961484432220459], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006447.pt"}
{"idx": 6449, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wj8XXvD4kGE.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where chocolate chips are being added to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where chocolate chips are being added to the mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 68.0], "base_se_norm": [-0.468999445438385, -0.38799941539764404], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.468999445438385, -0.4419994354248047], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006449.pt"}
{"idx": 6451, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wj8XXvD4kGE.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the dough is cut into nine equal pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the dough is cut into nine equal pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 142.0], "base_se_norm": [0.15200114250183105, 0.2780013084411621], "gt_se_raw": [127.0, 142.0], "gt_se_norm": [0.14300119876861572, 0.2780013084411621], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006451.pt"}
{"idx": 6453, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M-H8mtPWeKw.mp4", "duration": 238.17126666666667, "prompt_text": "Please find the visual event described by the sentence 'When are the scallions and ginger added to the wok?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the scallions and ginger added to the wok?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 44.0], "base_se_norm": [-0.6557099223136902, -0.6305179595947266], "gt_se_raw": [41.0, 43.0], "gt_se_norm": [-0.6557099223136902, -0.6389153003692627], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006453.pt"}
{"idx": 6455, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M-H8mtPWeKw.mp4", "duration": 238.17126666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the prosciutto is added to the vegetables.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 113.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the prosciutto is added to the vegetables.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 113.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [113.0, 129.0], "base_se_norm": [-0.05110299587249756, 0.08325409889221191], "gt_se_raw": [113.0, 115.0], "gt_se_norm": [-0.05110299587249756, -0.03430837392807007], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006455.pt"}
{"idx": 6457, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/M-H8mtPWeKw.mp4", "duration": 238.17126666666667, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the eggs are cracked into the wok.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 161.0 - 164.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the eggs are cracked into the wok.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 161.0 - 164.0 seconds.<|im_end|>\n", "base_se_raw": [161.0, 164.0], "base_se_norm": [0.35196828842163086, 0.37716031074523926], "gt_se_raw": [161.0, 163.0], "gt_se_norm": [0.35196828842163086, 0.3687629699707031], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006457.pt"}
{"idx": 6459, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7gquYRxLMFI.mp4", "duration": 205.45524999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point is a blue tractor driving through the vineyard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a blue tractor driving through the vineyard?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 28.0], "base_se_norm": [-0.7566379904747009, -0.727434515953064], "gt_se_raw": [25.0, 27.0], "gt_se_norm": [-0.7566379904747009, -0.7371690273284912], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006459.pt"}
{"idx": 6461, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7gquYRxLMFI.mp4", "duration": 205.45524999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where the juice is being pressed and extracted.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where the juice is being pressed and extracted.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 121.0], "base_se_norm": [0.0707927942276001, 0.17787206172943115], "gt_se_raw": [114.0, 118.0], "gt_se_norm": [0.10973072052001953, 0.14866864681243896], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006461.pt"}
{"idx": 6463, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uTDkM7wUx-A.mp4", "duration": 278.0, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when a man in uniform opens a large wooden door and enters a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 4.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when a man in uniform opens a large wooden door and enters a room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 4.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 4.0], "base_se_norm": [-0.9928057789802551, -0.971222996711731], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9928057789802551, -0.971222996711731], "start_token_span": [72, 75], "end_token_span": [77, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006463.pt"}
{"idx": 6465, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uTDkM7wUx-A.mp4", "duration": 278.0, "prompt_text": "Please find the visual event described by the sentence 'At what point does the older woman in the black hat and fur collar argue face-to-face with the man in the leather coat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the older woman in the black hat and fur collar argue face-to-face with the man in the leather coat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 182.0], "base_se_norm": [0.1798560619354248, 0.3093525171279907], "gt_se_raw": [168.0, 181.0], "gt_se_norm": [0.20863306522369385, 0.3021582365036011], "start_token_span": [80, 85], "end_token_span": [87, 92], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006465.pt"}
{"idx": 6467, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/uTDkM7wUx-A.mp4", "duration": 278.0, "prompt_text": "Please find the visual event described by the sentence 'What is the time range when the man at the desk is speaking to himself after the uniformed man has left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time range when the man at the desk is speaking to himself after the uniformed man has left?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 79.0], "base_se_norm": [-0.6618704795837402, -0.431654691696167], "gt_se_raw": [51.0, 70.0], "gt_se_norm": [-0.633093535900116, -0.4964028596878052], "start_token_span": [77, 81], "end_token_span": [83, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006467.pt"}
{"idx": 6469, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TWfph3iNC-k.mp4", "duration": 209.1089, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the person jumps off the cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the person jumps off the cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.6939393877983093, -0.6556817889213562], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.6939393877983093, -0.6748105883598328], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006469.pt"}
{"idx": 6471, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TWfph3iNC-k.mp4", "duration": 209.1089, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for the flyer deploying his parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 134.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for the flyer deploying his parachute?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 134.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 134.0], "base_se_norm": [0.21467804908752441, 0.2816288471221924], "gt_se_raw": [127.0, 129.0], "gt_se_norm": [0.21467804908752441, 0.23380684852600098], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006471.pt"}
{"idx": 6473, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yjikAEjCPEI.mp4", "duration": 282.8158666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man begin to layer the first vocal melody over the beat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man begin to layer the first vocal melody over the beat?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 144.0], "base_se_norm": [-0.13017606735229492, 0.01833045482635498], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [-0.7524890899658203, -0.7100586891174316], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006473.pt"}
{"idx": 6475, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yjikAEjCPEI.mp4", "duration": 282.8158666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man starts beatboxing to create the main drum rhythm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man starts beatboxing to create the main drum rhythm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 51.0], "base_se_norm": [-0.7524890899658203, -0.6393412947654724], "gt_se_raw": [6.0, 12.0], "gt_se_norm": [-0.9575695395469666, -0.9151391386985779], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006475.pt"}
{"idx": 6477, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yjikAEjCPEI.mp4", "duration": 282.8158666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man start repeating the lyric 'We're gonna gain control'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 123.0 - 144.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man start repeating the lyric 'We're gonna gain control'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 123.0 - 144.0 seconds.<|im_end|>\n", "base_se_raw": [123.0, 144.0], "base_se_norm": [-0.13017606735229492, 0.01833045482635498], "gt_se_raw": [233.0, 238.0], "gt_se_norm": [0.6477152109146118, 0.6830739974975586], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006477.pt"}
{"idx": 6479, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jzF_y039slk.mp4", "duration": 239.69779166666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the shot where the man is playing the cello on the edge of a cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the shot where the man is playing the cello on the edge of a cliff.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.7329971194267273, -0.6996217966079712], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.7329971194267273, -0.6996217966079712], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006479.pt"}
{"idx": 6481, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jzF_y039slk.mp4", "duration": 239.69779166666666, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the man is playing a black electric cello inside a cave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the man is playing a black electric cello inside a cave.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.6078395247459412, -0.5661203265190125], "gt_se_raw": [47.0, 49.0], "gt_se_norm": [-0.6078395247459412, -0.5911518335342407], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006481.pt"}
{"idx": 6483, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NLlGopyXT_g.mp4", "duration": 273.4, "prompt_text": "Please find the visual event described by the sentence 'When do we see someone writing 'Text is unilinear' on a piece of paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do we see someone writing 'Text is unilinear' on a piece of paper?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 13.0], "base_se_norm": [-0.9926847219467163, -0.9049012660980225], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.9707388281822205], "start_token_span": [72, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006483.pt"}
{"idx": 6485, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NLlGopyXT_g.mp4", "duration": 273.4, "prompt_text": "Please find the visual event described by the sentence 'At what point is the CNN.com homepage from January 29, 2007, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the CNN.com homepage from January 29, 2007, displayed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 88.0], "base_se_norm": [-0.392831027507782, -0.35625457763671875], "gt_se_raw": [83.0, 86.0], "gt_se_norm": [-0.392831027507782, -0.37088513374328613], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006485.pt"}
{"idx": 6487, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/NLlGopyXT_g.mp4", "duration": 273.4, "prompt_text": "Please find the visual event described by the sentence 'When are the search results for 'anthropology 2.0 hypermedia' being tagged on del.icio.us?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 174.0 - 180.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the search results for 'anthropology 2.0 hypermedia' being tagged on del.icio.us?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 174.0 - 180.0 seconds.<|im_end|>\n", "base_se_raw": [174.0, 180.0], "base_se_norm": [0.2728602886199951, 0.3167520761489868], "gt_se_raw": [172.0, 179.0], "gt_se_norm": [0.25822973251342773, 0.3094367980957031], "start_token_span": [79, 84], "end_token_span": [86, 91], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006487.pt"}
{"idx": 6489, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fhZo17Pxq1A.mp4", "duration": 207.207, "prompt_text": "Please find the visual event described by the sentence 'Show me the moment the Psyduck is rolling the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the moment the Psyduck is rolling the dough.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 60.0], "base_se_norm": [-0.49808645248413086, -0.42086899280548096], "gt_se_raw": [51.0, 58.0], "gt_se_norm": [-0.5077386498451233, -0.44017332792282104], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006489.pt"}
{"idx": 6491, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/fhZo17Pxq1A.mp4", "duration": 207.207, "prompt_text": "Please find the visual event described by the sentence 'Locate the time when the rabbit is spreading sauce on the pancake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time when the rabbit is spreading sauce on the pancake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 102.0], "base_se_norm": [-0.06373822689056396, -0.015477299690246582], "gt_se_raw": [97.0, 102.0], "gt_se_norm": [-0.06373822689056396, -0.015477299690246582], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006491.pt"}
{"idx": 6493, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t2mU6USTBRE.mp4", "duration": 295.12816666666663, "prompt_text": "Please find the visual event described by the sentence 'When is the man being confronted by three other men in a black and white scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 51.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man being confronted by three other men in a black and white scene?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 51.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 51.0], "base_se_norm": [-0.911902666091919, -0.6543874740600586], "gt_se_raw": [12.0, 65.0], "gt_se_norm": [-0.918679416179657, -0.5595134496688843], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006493.pt"}
{"idx": 6495, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t2mU6USTBRE.mp4", "duration": 295.12816666666663, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment when one of the dancers roller skates across a red line painted on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 272.0 - 275.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment when one of the dancers roller skates across a red line painted on the floor.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 272.0 - 275.0 seconds.<|im_end|>\n", "base_se_raw": [272.0, 275.0], "base_se_norm": [0.8432669639587402, 0.86359703540802], "gt_se_raw": [272.0, 275.0], "gt_se_norm": [0.8432669639587402, 0.86359703540802], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006495.pt"}
{"idx": 6497, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t2mU6USTBRE.mp4", "duration": 295.12816666666663, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man dressed in black is running up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 234.0 - 242.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man dressed in black is running up the stairs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 234.0 - 242.0 seconds.<|im_end|>\n", "base_se_raw": [234.0, 242.0], "base_se_norm": [0.5857516527175903, 0.6399654150009155], "gt_se_raw": [235.0, 239.0], "gt_se_norm": [0.5925284624099731, 0.6196352243423462], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006497.pt"}
{"idx": 6499, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mObyN915RkE.mp4", "duration": 205.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child pick up the 'puzzle' icon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child pick up the 'puzzle' icon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 65.0], "base_se_norm": [-0.44543540477752686, -0.36760175228118896], "gt_se_raw": [105.0, 107.0], "gt_se_norm": [0.021566390991210938, 0.04102480411529541], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006499.pt"}
{"idx": 6501, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mObyN915RkE.mp4", "duration": 205.56666666666666, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when the child touches the braille for the word 'puzzle'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when the child touches the braille for the word 'puzzle'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.688665509223938, -0.6108318567276001], "gt_se_raw": [16.0, 19.0], "gt_se_norm": [-0.844332754611969, -0.8151451349258423], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006501.pt"}
{"idx": 6503, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ewq5tStHmdk.mp4", "duration": 299.09045555555554, "prompt_text": "Please find the visual event described by the sentence 'When do the man and woman embrace each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the man and woman embrace each other?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 22.0], "base_se_norm": [-0.899695873260498, -0.8528873324394226], "gt_se_raw": [15.0, 21.0], "gt_se_norm": [-0.899695873260498, -0.8595742583274841], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006503.pt"}
{"idx": 6505, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ewq5tStHmdk.mp4", "duration": 299.09045555555554, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up shot of a woman's hand signing a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up shot of a woman's hand signing a document.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 96.0], "base_se_norm": [-0.45167088508605957, -0.3580537438392639], "gt_se_raw": [83.0, 86.0], "gt_se_norm": [-0.44498395919799805, -0.4249231219291687], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006505.pt"}
{"idx": 6507, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ewq5tStHmdk.mp4", "duration": 299.09045555555554, "prompt_text": "Please find the visual event described by the sentence 'When does the woman sarcastically explain to the waitress that the man is in love with his laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 267.0 - 283.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman sarcastically explain to the waitress that the man is in love with his laptop?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 267.0 - 283.0 seconds.<|im_end|>\n", "base_se_raw": [267.0, 283.0], "base_se_norm": [0.7854130268096924, 0.8924040794372559], "gt_se_raw": [269.0, 274.0], "gt_se_norm": [0.7987868785858154, 0.8322216272354126], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006507.pt"}
{"idx": 6509, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZREM_AcXHX8.mp4", "duration": 227.5273, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is talking about the trajectory of the arrow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 185.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is talking about the trajectory of the arrow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 185.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 185.0], "base_se_norm": [0.45037543773651123, 0.626178503036499], "gt_se_raw": [167.0, 185.0], "gt_se_norm": [0.46795570850372314, 0.626178503036499], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006509.pt"}
{"idx": 6511, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ZREM_AcXHX8.mp4", "duration": 227.5273, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman explains the role of feathers and tips in stabilizing arrows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 89.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman explains the role of feathers and tips in stabilizing arrows.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 89.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [89.0, 105.0], "base_se_norm": [-0.2176762819290161, -0.07703381776809692], "gt_se_raw": [90.0, 105.0], "gt_se_norm": [-0.20888614654541016, -0.07703381776809692], "start_token_span": [71, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006511.pt"}
{"idx": 6513, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/07d2dXHYb94.mp4", "duration": 245.0, "prompt_text": "Please find the visual event described by the sentence 'At what point is the white puppy watching as the other puppies run past?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 18.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the white puppy watching as the other puppies run past?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 18.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 18.0], "base_se_norm": [-0.9020408391952515, -0.8530611991882324], "gt_se_raw": [13.0, 17.0], "gt_se_norm": [-0.8938775658607483, -0.8612244725227356], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006513.pt"}
{"idx": 6515, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/07d2dXHYb94.mp4", "duration": 245.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the puppy causes the food dispenser to overflow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the puppy causes the food dispenser to overflow.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 78.0], "base_se_norm": [-0.40408164262771606, -0.36326533555984497], "gt_se_raw": [73.0, 78.0], "gt_se_norm": [-0.40408164262771606, -0.36326533555984497], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006515.pt"}
{"idx": 6517, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/07d2dXHYb94.mp4", "duration": 245.0, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the man put a blue cape on the white puppy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 208.0 - 211.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the man put a blue cape on the white puppy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 208.0 - 211.0 seconds.<|im_end|>\n", "base_se_raw": [208.0, 211.0], "base_se_norm": [0.6979591846466064, 0.7224489450454712], "gt_se_raw": [208.0, 210.0], "gt_se_norm": [0.6979591846466064, 0.7142857313156128], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006517.pt"}
{"idx": 6519, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u6euUDxXlAM.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man presents the 'BE A BROADWAY STAR!' board game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 162.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man presents the 'BE A BROADWAY STAR!' board game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 162.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 162.0], "base_se_norm": [-0.15058857202529907, -0.01710963249206543], "gt_se_raw": [140.0, 145.0], "gt_se_norm": [-0.15058857202529907, -0.12025243043899536], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006519.pt"}
{"idx": 6521, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/u6euUDxXlAM.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman sings a very high-pitched note.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 293.0 - 296.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman sings a very high-pitched note.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 293.0 - 296.0 seconds.<|im_end|>\n", "base_se_raw": [293.0, 296.0], "base_se_norm": [0.7776968479156494, 0.7958984375], "gt_se_raw": [239.0, 242.0], "gt_se_norm": [0.4500666856765747, 0.46826839447021484], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006521.pt"}
{"idx": 6523, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggt_6Xb0c48.mp4", "duration": 301.80150000000003, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the woman demonstrates how to make a DIY BB cream.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 74.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the woman demonstrates how to make a DIY BB cream.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 74.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [74.0, 93.0], "base_se_norm": [-0.5096114873886108, -0.3837009072303772], "gt_se_raw": [75.0, 87.0], "gt_se_norm": [-0.5029846429824829, -0.4234621524810791], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006523.pt"}
{"idx": 6525, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggt_6Xb0c48.mp4", "duration": 301.80150000000003, "prompt_text": "Please find the visual event described by the sentence 'When is the woman applying concealer under her eyes with a wand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the woman applying concealer under her eyes with a wand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 145.0], "base_se_norm": [-0.09211856126785278, -0.039103567600250244], "gt_se_raw": [137.0, 145.0], "gt_se_norm": [-0.09211856126785278, -0.039103567600250244], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006525.pt"}
{"idx": 6527, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ggt_6Xb0c48.mp4", "duration": 301.80150000000003, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the woman is putting on lipstick pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 249.0 - 259.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the woman is putting on lipstick pencil.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 249.0 - 259.0 seconds.<|im_end|>\n", "base_se_raw": [249.0, 259.0], "base_se_norm": [0.6500911712646484, 0.7163598537445068], "gt_se_raw": [250.0, 259.0], "gt_se_norm": [0.6567180156707764, 0.7163598537445068], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006527.pt"}
{"idx": 6529, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BxyXk1sr2io.mp4", "duration": 251.08, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the four corners are folded towards the center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 96.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the four corners are folded towards the center.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 96.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 96.0], "base_se_norm": [-0.6813764572143555, -0.23530352115631104], "gt_se_raw": [71.0, 96.0], "gt_se_norm": [-0.43444323539733887, -0.23530352115631104], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006529.pt"}
{"idx": 6531, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BxyXk1sr2io.mp4", "duration": 251.08, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where numbers are written on the inner triangles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where numbers are written on the inner triangles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 157.0], "base_se_norm": [0.13907921314239502, 0.25059735774993896], "gt_se_raw": [143.0, 156.0], "gt_se_norm": [0.13907921314239502, 0.24263179302215576], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006531.pt"}
{"idx": 6533, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RLmKfXwWQtE.mp4", "duration": 301.42612499999996, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video introduce the concept of the population pyramid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video introduce the concept of the population pyramid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 88.0], "base_se_norm": [-0.6217978596687317, -0.41610896587371826], "gt_se_raw": [57.0, 60.0], "gt_se_norm": [-0.6217978596687317, -0.6018924713134766], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006533.pt"}
{"idx": 6535, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RLmKfXwWQtE.mp4", "duration": 301.42612499999996, "prompt_text": "Please find the visual event described by the sentence 'When is the population growth chart for the entire globe shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the population growth chart for the entire globe shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 41.0], "base_se_norm": [-0.8208516240119934, -0.7279598712921143], "gt_se_raw": [30.0, 36.0], "gt_se_norm": [-0.8009462356567383, -0.761135458946228], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006535.pt"}
{"idx": 6537, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/RLmKfXwWQtE.mp4", "duration": 301.42612499999996, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the video explains the effect of China's one-child policy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 238.0 - 253.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the video explains the effect of China's one-child policy.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 238.0 - 253.0 seconds.<|im_end|>\n", "base_se_raw": [238.0, 253.0], "base_se_norm": [0.5791597366333008, 0.6786866188049316], "gt_se_raw": [238.0, 253.0], "gt_se_norm": [0.5791597366333008, 0.6786866188049316], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006537.pt"}
{"idx": 6539, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E1dVEAdYuaM.mp4", "duration": 261.6, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the two women are discussing the arrival of the package.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the two women are discussing the arrival of the package.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 48.0], "base_se_norm": [-0.7706422209739685, -0.6330275535583496], "gt_se_raw": [31.0, 48.0], "gt_se_norm": [-0.7629969120025635, -0.6330275535583496], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006539.pt"}
{"idx": 6541, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/E1dVEAdYuaM.mp4", "duration": 261.6, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment the man asks the woman to marry him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 184.0 - 216.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment the man asks the woman to marry him.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 184.0 - 216.0 seconds.<|im_end|>\n", "base_se_raw": [184.0, 216.0], "base_se_norm": [0.40672779083251953, 0.6513761281967163], "gt_se_raw": [224.0, 230.0], "gt_se_norm": [0.7125382423400879, 0.7584097385406494], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006541.pt"}
{"idx": 6543, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ta9K22D0o5Q.mp4", "duration": 251.4178333333333, "prompt_text": "Please find the visual event described by the sentence 'When can we see the shell climbing on a pile of papers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the shell climbing on a pile of papers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 71.0], "base_se_norm": [-0.4988422393798828, -0.4352031350135803], "gt_se_raw": [68.0, 70.0], "gt_se_norm": [-0.45906782150268555, -0.44315803050994873], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006543.pt"}
{"idx": 6545, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ta9K22D0o5Q.mp4", "duration": 251.4178333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where the shell is sitting on a laptop, talking about its sister.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 219.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where the shell is sitting on a laptop, talking about its sister.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 219.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 219.0], "base_se_norm": [-0.005639314651489258, 0.7421199083328247], "gt_se_raw": [129.0, 140.0], "gt_se_norm": [0.026180148124694824, 0.11368393898010254], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006545.pt"}
{"idx": 6547, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Ta9K22D0o5Q.mp4", "duration": 251.4178333333333, "prompt_text": "Please find the visual event described by the sentence 'Where is the shell sitting in a blue fruit bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 115.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where is the shell sitting in a blue fruit bowl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 115.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 115.0], "base_se_norm": [-0.25224077701568604, -0.08518821001052856], "gt_se_raw": [94.0, 103.0], "gt_se_norm": [-0.25224077701568604, -0.1806468367576599], "start_token_span": [65, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006547.pt"}
{"idx": 6549, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ka8-Nefp_gk.mp4", "duration": 297.5833333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the words 'Inclusive~Included~Inclusion' appear at the top of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the words 'Inclusive~Included~Inclusion' appear at the top of the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 109.0], "base_se_norm": [-0.6505180597305298, -0.2674320936203003], "gt_se_raw": [52.0, 110.0], "gt_se_norm": [-0.6505180597305298, -0.26071131229400635], "start_token_span": [76, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006549.pt"}
{"idx": 6551, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ka8-Nefp_gk.mp4", "duration": 297.5833333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a piggy bank shown with money going into it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 189.0 - 196.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a piggy bank shown with money going into it?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 189.0 - 196.0 seconds.<|im_end|>\n", "base_se_raw": [189.0, 196.0], "base_se_norm": [0.2702324390411377, 0.31727802753448486], "gt_se_raw": [189.0, 194.0], "gt_se_norm": [0.2702324390411377, 0.303836464881897], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006551.pt"}
{"idx": 6553, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/phQDinMbmic.mp4", "duration": 219.8863333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment a pencil draws a line on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment a pencil draws a line on a piece of paper.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 10.0], "base_se_norm": [-0.9545219540596008, -0.9090439081192017], "gt_se_raw": [8.0, 10.0], "gt_se_norm": [-0.9272351264953613, -0.9090439081192017], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006553.pt"}
{"idx": 6555, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/phQDinMbmic.mp4", "duration": 219.8863333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the boy give the drawing back to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the boy give the drawing back to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 145.0], "base_se_norm": [0.2642894983291626, 0.3188631534576416], "gt_se_raw": [139.0, 142.0], "gt_se_norm": [0.2642894983291626, 0.2915763854980469], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006555.pt"}
{"idx": 6557, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/phQDinMbmic.mp4", "duration": 219.8863333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the boy sees the drawing of him and Alice together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 184.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the boy sees the drawing of him and Alice together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 184.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 184.0], "base_se_norm": [0.6463050842285156, 0.6735919713973999], "gt_se_raw": [180.0, 184.0], "gt_se_norm": [0.6372095346450806, 0.6735919713973999], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006557.pt"}
{"idx": 6559, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5GN90UiImo.mp4", "duration": 314.88123333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man in the pink tie get thrown out of the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 160.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man in the pink tie get thrown out of the window?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 160.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 160.0], "base_se_norm": [-0.03455662727355957, 0.016256213188171387], "gt_se_raw": [157.0, 159.0], "gt_se_norm": [-0.002798616886138916, 0.009904623031616211], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006559.pt"}
{"idx": 6561, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/d5GN90UiImo.mp4", "duration": 314.88123333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman in the red dress turns into She-Rock Obama.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 252.0 - 277.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman in the red dress turns into She-Rock Obama.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 252.0 - 277.0 seconds.<|im_end|>\n", "base_se_raw": [252.0, 277.0], "base_se_norm": [0.6006035804748535, 0.759393572807312], "gt_se_raw": [252.0, 266.0], "gt_se_norm": [0.6006035804748535, 0.6895259618759155], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006561.pt"}
{"idx": 6563, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MoOXcFZgzhI.mp4", "duration": 218.05116666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a girl uses a stick to draw a circle on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 17.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a girl uses a stick to draw a circle on the grass.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 17.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [17.0, 43.0], "base_se_norm": [-0.8440732955932617, -0.6055971384048462], "gt_se_raw": [37.0, 42.0], "gt_se_norm": [-0.6606301069259644, -0.6147693395614624], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006563.pt"}
{"idx": 6565, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MoOXcFZgzhI.mp4", "duration": 218.05116666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the red vest appear and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 136.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the red vest appear and speak?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 136.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 136.0], "base_se_norm": [0.008937478065490723, 0.24741363525390625], "gt_se_raw": [110.0, 136.0], "gt_se_norm": [0.008937478065490723, 0.24741363525390625], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006565.pt"}
{"idx": 6567, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/MoOXcFZgzhI.mp4", "duration": 218.05116666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where a man in a suit and tie is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 187.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where a man in a suit and tie is speaking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 187.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 187.0], "base_se_norm": [0.4675455093383789, 0.7151937484741211], "gt_se_raw": [160.0, 186.0], "gt_se_norm": [0.4675455093383789, 0.7060216665267944], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006567.pt"}
{"idx": 6569, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CN8899eqays.mp4", "duration": 311.97833333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the woman apply foundation with a brush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 47.0 - 52.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the woman apply foundation with a brush?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 47.0 - 52.0 seconds.<|im_end|>\n", "base_se_raw": [47.0, 52.0], "base_se_norm": [-0.698697030544281, -0.666643500328064], "gt_se_raw": [48.0, 52.0], "gt_se_norm": [-0.6922863125801086, -0.666643500328064], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006569.pt"}
{"idx": 6571, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/CN8899eqays.mp4", "duration": 311.97833333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where she applies layers of tissue paper and latex to her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 102.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where she applies layers of tissue paper and latex to her face.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 102.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [102.0, 141.0], "base_se_norm": [-0.34610843658447266, -0.09609109163284302], "gt_se_raw": [105.0, 121.0], "gt_se_norm": [-0.32687634229660034, -0.22430509328842163], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006571.pt"}
{"idx": 6573, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKGeJXk2qWQ.mp4", "duration": 234.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the cashier is waiting alone and looks bored after the first customer leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the cashier is waiting alone and looks bored after the first customer leaves.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 38.0], "base_se_norm": [-0.8466725945472717, -0.6763087511062622], "gt_se_raw": [18.0, 37.0], "gt_se_norm": [-0.8466725945472717, -0.6848269701004028], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006573.pt"}
{"idx": 6575, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKGeJXk2qWQ.mp4", "duration": 234.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the cashier and the customer leave the store and walk into the desert together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the cashier and the customer leave the store and walk into the desert together.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 155.0], "base_se_norm": [0.2181011438369751, 0.32031941413879395], "gt_se_raw": [143.0, 155.0], "gt_se_norm": [0.2181011438369751, 0.32031941413879395], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006575.pt"}
{"idx": 6577, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/yKGeJXk2qWQ.mp4", "duration": 234.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'When does the customer reveal that the creature in the hole loves butter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 163.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the customer reveal that the creature in the hole loves butter?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 163.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [163.0, 172.0], "base_se_norm": [0.38846492767333984, 0.46512866020202637], "gt_se_raw": [190.0, 194.0], "gt_se_norm": [0.6184560060501099, 0.6525287628173828], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006577.pt"}
{"idx": 6579, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VF9-sEbqDvU.mp4", "duration": 201.61808333333332, "prompt_text": "Please find the visual event described by the sentence 'At what point does the shell attempt to lift a crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 151.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the shell attempt to lift a crayon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 151.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [151.0, 158.0], "base_se_norm": [0.49788153171539307, 0.5673197507858276], "gt_se_raw": [152.0, 161.0], "gt_se_norm": [0.5078011751174927, 0.5970789194107056], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006579.pt"}
{"idx": 6581, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/VF9-sEbqDvU.mp4", "duration": 201.61808333333332, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where the shell is on the kitchen counter next to a Coca-Cola bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 30.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where the shell is on the kitchen counter next to a Coca-Cola bottle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 30.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [30.0, 43.0], "base_se_norm": [-0.7024076581001282, -0.5734509825706482], "gt_se_raw": [29.0, 43.0], "gt_se_norm": [-0.7123273611068726, -0.5734509825706482], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006581.pt"}
{"idx": 6583, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BA2yyhbHcCc.mp4", "duration": 309.20920920920923, "prompt_text": "Please find the visual event described by the sentence 'When does the camera switch to a first-person point of view, showing the person's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera switch to a first-person point of view, showing the person's hands?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 30.0], "base_se_norm": [-0.8382971882820129, -0.8059566020965576], "gt_se_raw": [26.0, 29.0], "gt_se_norm": [-0.8318290710449219, -0.8124247193336487], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006583.pt"}
{"idx": 6585, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BA2yyhbHcCc.mp4", "duration": 309.20920920920923, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where we see parasites that look like two flatworms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 198.0 - 202.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where we see parasites that look like two flatworms.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 198.0 - 202.0 seconds.<|im_end|>\n", "base_se_raw": [198.0, 202.0], "base_se_norm": [0.2806863784790039, 0.30655884742736816], "gt_se_raw": [198.0, 202.0], "gt_se_norm": [0.2806863784790039, 0.30655884742736816], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006585.pt"}
{"idx": 6587, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/BA2yyhbHcCc.mp4", "duration": 309.20920920920923, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman in a black vest is looking into a compound microscope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 169.0 - 184.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman in a black vest is looking into a compound microscope.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 169.0 - 184.0 seconds.<|im_end|>\n", "base_se_raw": [169.0, 184.0], "base_se_norm": [0.09311103820800781, 0.190132737159729], "gt_se_raw": [169.0, 171.0], "gt_se_norm": [0.09311103820800781, 0.10604727268218994], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006587.pt"}
{"idx": 6589, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oRmpxPwJFw8.mp4", "duration": 222.55566666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me where the two forks are placed on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 64.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me where the two forks are placed on the table.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 64.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 64.0], "base_se_norm": [-0.4697955846786499, -0.42486298084259033], "gt_se_raw": [51.0, 63.0], "gt_se_norm": [-0.5416877269744873, -0.4338495135307312], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006589.pt"}
{"idx": 6591, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/oRmpxPwJFw8.mp4", "duration": 222.55566666666664, "prompt_text": "Please find the visual event described by the sentence 'Find the segment where the presenter demonstrates the 'BMW' tip, referring to bread, meal, and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the segment where the presenter demonstrates the 'BMW' tip, referring to bread, meal, and water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 173.0], "base_se_norm": [0.4378424882888794, 0.5546672344207764], "gt_se_raw": [146.0, 154.0], "gt_se_norm": [0.31203126907348633, 0.38392341136932373], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006591.pt"}
{"idx": 6593, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wx9vPv-T51I.mp4", "duration": 216.5079583333333, "prompt_text": "Please find the visual event described by the sentence 'Where can we see the soldiers running on the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can we see the soldiers running on the beach?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 10.0], "base_se_norm": [-0.944574773311615, -0.9076246619224548], "gt_se_raw": [6.0, 9.0], "gt_se_norm": [-0.944574773311615, -0.9168621897697449], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006593.pt"}
{"idx": 6595, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wx9vPv-T51I.mp4", "duration": 216.5079583333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see the soldiers celebrating their victory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see the soldiers celebrating their victory?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 32.0], "base_se_norm": [-0.7321114540100098, -0.7043988704681396], "gt_se_raw": [30.0, 32.0], "gt_se_norm": [-0.7228739261627197, -0.7043988704681396], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006595.pt"}
{"idx": 6597, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wx9vPv-T51I.mp4", "duration": 216.5079583333333, "prompt_text": "Please find the visual event described by the sentence 'When does the animation show the tectonic plates shifting on the ocean floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 77.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the animation show the tectonic plates shifting on the ocean floor?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 77.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 77.0], "base_se_norm": [-0.37184756994247437, -0.28870975971221924], "gt_se_raw": [75.0, 78.0], "gt_se_norm": [-0.3071848154067993, -0.2794722318649292], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006597.pt"}
{"idx": 6599, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xiSIQzwIPzQ.mp4", "duration": 210.2, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a boy is excited to see the pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 8.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a boy is excited to see the pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 8.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 8.0], "base_se_norm": [-0.9524262547492981, -0.923882007598877], "gt_se_raw": [49.0, 53.0], "gt_se_norm": [-0.5337773561477661, -0.4957183599472046], "start_token_span": [67, 70], "end_token_span": [72, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006599.pt"}
{"idx": 6601, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/xiSIQzwIPzQ.mp4", "duration": 210.2, "prompt_text": "Please find the visual event described by the sentence 'When does the girl in the blue dress get angry and yell at the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl in the blue dress get angry and yell at the boy?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 111.0], "base_se_norm": [-0.36251187324523926, 0.0561370849609375], "gt_se_raw": [67.0, 81.0], "gt_se_norm": [-0.36251187324523926, -0.22930538654327393], "start_token_span": [70, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006601.pt"}
{"idx": 6603, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3X1nPsafKdk.mp4", "duration": 208.0, "prompt_text": "Please find the visual event described by the sentence 'When can we see the host shaking hands with the puppet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 10.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the host shaking hands with the puppet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 10.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 10.0], "base_se_norm": [-0.932692289352417, -0.9038461446762085], "gt_se_raw": [7.0, 9.0], "gt_se_norm": [-0.932692289352417, -0.9134615659713745], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006603.pt"}
{"idx": 6605, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3X1nPsafKdk.mp4", "duration": 208.0, "prompt_text": "Please find the visual event described by the sentence 'When does the host show her own puppet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 97.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the host show her own puppet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 97.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [97.0, 105.0], "base_se_norm": [-0.06730771064758301, 0.009615421295166016], "gt_se_raw": [94.0, 98.0], "gt_se_norm": [-0.0961538553237915, -0.05769228935241699], "start_token_span": [63, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006605.pt"}
{"idx": 6607, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3X1nPsafKdk.mp4", "duration": 208.0, "prompt_text": "Please find the visual event described by the sentence 'When does the girl put the host's puppet on the hay?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 155.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the girl put the host's puppet on the hay?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 155.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [155.0, 158.0], "base_se_norm": [0.490384578704834, 0.5192307233810425], "gt_se_raw": [175.0, 179.0], "gt_se_norm": [0.682692289352417, 0.7211538553237915], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006607.pt"}
{"idx": 6609, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJJBHDi9axY.mp4", "duration": 324.69103333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when two people are sticking out of trash cans upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 66.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when two people are sticking out of trash cans upside down.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 66.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 66.0], "base_se_norm": [-0.6365776062011719, -0.5934596657752991], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.6365776062011719, -0.5996193885803223], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006609.pt"}
{"idx": 6611, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pJJBHDi9axY.mp4", "duration": 324.69103333333334, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a woman gets shot with a puff of smoke and collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a woman gets shot with a puff of smoke and collapses.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 76.0], "base_se_norm": [-0.5996193885803223, -0.5318626165390015], "gt_se_raw": [231.0, 234.0], "gt_se_norm": [0.42289113998413086, 0.44137024879455566], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006611.pt"}
{"idx": 6613, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2is08NsKzkw.mp4", "duration": 256.7565, "prompt_text": "Please find the visual event described by the sentence 'When can we see the text 'Yellowstone National Park' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 20.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the text 'Yellowstone National Park' displayed on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 20.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [20.0, 28.0], "base_se_norm": [-0.8442103862762451, -0.7818945050239563], "gt_se_raw": [20.0, 26.0], "gt_se_norm": [-0.8442103862762451, -0.7974734902381897], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006613.pt"}
{"idx": 6615, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2is08NsKzkw.mp4", "duration": 256.7565, "prompt_text": "Please find the visual event described by the sentence 'At what point does a park ranger use an infrared camera to measure the ground temperature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a park ranger use an infrared camera to measure the ground temperature?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 155.0], "base_se_norm": [0.12168538570404053, 0.20736968517303467], "gt_se_raw": [145.0, 155.0], "gt_se_norm": [0.12947475910186768, 0.20736968517303467], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006615.pt"}
{"idx": 6617, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2is08NsKzkw.mp4", "duration": 256.7565, "prompt_text": "Please find the visual event described by the sentence 'Find the scene showing a herd of bison grazing on a grassy field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 236.0 - 240.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene showing a herd of bison grazing on a grassy field.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 236.0 - 240.0 seconds.<|im_end|>\n", "base_se_raw": [236.0, 240.0], "base_se_norm": [0.8383176326751709, 0.8694756031036377], "gt_se_raw": [236.0, 240.0], "gt_se_norm": [0.8383176326751709, 0.8694756031036377], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006617.pt"}
{"idx": 6619, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFtBjc1dz7w.mp4", "duration": 290.2482888888889, "prompt_text": "Please find the visual event described by the sentence 'At what point does the narrator jokingly refer to the Elephant Graveyard as 'New Jersey'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the narrator jokingly refer to the Elephant Graveyard as 'New Jersey'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 28.0], "base_se_norm": [-0.8208430409431458, -0.8070617318153381], "gt_se_raw": [42.0, 43.0], "gt_se_norm": [-0.7105926275253296, -0.7037019729614258], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006619.pt"}
{"idx": 6621, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DFtBjc1dz7w.mp4", "duration": 290.2482888888889, "prompt_text": "Please find the visual event described by the sentence 'When are the humorous nicknames for the characters, like 'Lion Jafar,' displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 230.0 - 242.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the humorous nicknames for the characters, like 'Lion Jafar,' displayed on screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 230.0 - 242.0 seconds.<|im_end|>\n", "base_se_raw": [230.0, 242.0], "base_se_norm": [0.5848499536514282, 0.6675378084182739], "gt_se_raw": [230.0, 242.0], "gt_se_norm": [0.5848499536514282, 0.6675378084182739], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006621.pt"}
{"idx": 6623, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCsVT9TKahk.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the woman state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 1.0 - 5.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman state her name?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 1.0 - 5.0 seconds.<|im_end|>\n", "base_se_raw": [1.0, 5.0], "base_se_norm": [-0.9900431632995605, -0.950215756893158], "gt_se_raw": [1.0, 3.0], "gt_se_norm": [-0.9900431632995605, -0.9701294302940369], "start_token_span": [62, 65], "end_token_span": [67, 70], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006623.pt"}
{"idx": 6625, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCsVT9TKahk.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does she pan the camera up to show the lights on her ceiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 131.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does she pan the camera up to show the lights on her ceiling?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 131.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [131.0, 137.0], "base_se_norm": [0.3043477535247803, 0.36408889293670654], "gt_se_raw": [131.0, 137.0], "gt_se_norm": [0.3043477535247803, 0.36408889293670654], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006625.pt"}
{"idx": 6627, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hCsVT9TKahk.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where she recommends earplugs for loud places.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 129.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where she recommends earplugs for loud places.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 129.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 129.0], "base_se_norm": [-0.0043146610260009766, 0.28443408012390137], "gt_se_raw": [81.0, 93.0], "gt_se_norm": [-0.19349485635757446, -0.0740126371383667], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006627.pt"}
{"idx": 6629, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_TbWcdN-W8o.mp4", "duration": 201.93506666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man reaches into a high cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 59.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man reaches into a high cabinet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 59.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [59.0, 65.0], "base_se_norm": [-0.4156537652015686, -0.35622870922088623], "gt_se_raw": [59.0, 65.0], "gt_se_norm": [-0.4156537652015686, -0.35622870922088623], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006629.pt"}
{"idx": 6631, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/_TbWcdN-W8o.mp4", "duration": 201.93506666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the man is playing bocce ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 138.0 - 141.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the man is playing bocce ball.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 138.0 - 141.0 seconds.<|im_end|>\n", "base_se_raw": [138.0, 141.0], "base_se_norm": [0.3667759895324707, 0.3964884281158447], "gt_se_raw": [138.0, 141.0], "gt_se_norm": [0.3667759895324707, 0.3964884281158447], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006631.pt"}
{"idx": 6633, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AW0jm6i9U3M.mp4", "duration": 298.46483333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the black car is driving alongside a body of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 2.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the black car is driving alongside a body of water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 2.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 2.0], "base_se_norm": [-1.0, -0.9865980744361877], "gt_se_raw": [0.0, 2.0], "gt_se_norm": [-1.0, -0.9865980744361877], "start_token_span": [68, 71], "end_token_span": [73, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006633.pt"}
{"idx": 6635, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AW0jm6i9U3M.mp4", "duration": 298.46483333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment the driver pulls a gun on the passenger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment the driver pulls a gun on the passenger.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 72.0], "base_se_norm": [-0.5644378066062927, -0.517531156539917], "gt_se_raw": [64.0, 70.0], "gt_se_norm": [-0.5711387395858765, -0.5309330224990845], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006635.pt"}
{"idx": 6637, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/AW0jm6i9U3M.mp4", "duration": 298.46483333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the passenger is coughing heavily after taking a hit from the pipe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the passenger is coughing heavily after taking a hit from the pipe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 114.0], "base_se_norm": [-0.269595742225647, -0.23609095811843872], "gt_se_raw": [113.0, 117.0], "gt_se_norm": [-0.24279189109802246, -0.21598803997039795], "start_token_span": [73, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006637.pt"}
{"idx": 6639, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zucl7HSA1M.mp4", "duration": 252.48556666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when a cat gets its head stuck in a plastic bag and runs off.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 46.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when a cat gets its head stuck in a plastic bag and runs off.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 46.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 46.0], "base_se_norm": [-0.6752289533615112, -0.6356227397918701], "gt_se_raw": [42.0, 46.0], "gt_se_norm": [-0.6673077344894409, -0.6356227397918701], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006639.pt"}
{"idx": 6641, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/3zucl7HSA1M.mp4", "duration": 252.48556666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where a cat jumps through a wall of toilet paper rolls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 112.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where a cat jumps through a wall of toilet paper rolls.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 112.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [112.0, 119.0], "base_se_norm": [-0.112820565700531, -0.05737185478210449], "gt_se_raw": [117.0, 118.0], "gt_se_norm": [-0.07321435213088989, -0.0652930736541748], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006641.pt"}
{"idx": 6643, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zWHu95io9B4.mp4", "duration": 219.63608333333332, "prompt_text": "Please find the visual event described by the sentence 'When are the breadcrumbs poured over the ground beef?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the breadcrumbs poured over the ground beef?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 27.0], "base_se_norm": [-0.7814566493034363, -0.7541387677192688], "gt_se_raw": [24.0, 27.0], "gt_se_norm": [-0.7814566493034363, -0.7541387677192688], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006643.pt"}
{"idx": 6645, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zWHu95io9B4.mp4", "duration": 219.63608333333332, "prompt_text": "Please find the visual event described by the sentence 'At what time are the meatballs shaped by hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time are the meatballs shaped by hand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 88.0], "base_se_norm": [-0.3625819683074951, -0.19867444038391113], "gt_se_raw": [81.0, 88.0], "gt_se_norm": [-0.2624162435531616, -0.19867444038391113], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006645.pt"}
{"idx": 6647, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zWHu95io9B4.mp4", "duration": 219.63608333333332, "prompt_text": "Please find the visual event described by the sentence 'Show me when the raw meatballs are carefully placed into the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 190.0 - 201.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the raw meatballs are carefully placed into the sauce.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 190.0 - 201.0 seconds.<|im_end|>\n", "base_se_raw": [190.0, 201.0], "base_se_norm": [0.7301347255706787, 0.8303004503250122], "gt_se_raw": [191.0, 199.0], "gt_se_norm": [0.7392406463623047, 0.8120884895324707], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006647.pt"}
{"idx": 6649, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UuAUKwAhIjY.mp4", "duration": 233.06616666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the midwife arrive at the house at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 67.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the midwife arrive at the house at night?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 67.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 67.0], "base_se_norm": [-0.519449770450592, -0.42505598068237305], "gt_se_raw": [56.0, 66.0], "gt_se_norm": [-0.519449770450592, -0.43363720178604126], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006649.pt"}
{"idx": 6651, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/UuAUKwAhIjY.mp4", "duration": 233.06616666666667, "prompt_text": "Please find the visual event described by the sentence 'At what point does the mother hold her newborn baby for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 122.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the mother hold her newborn baby for the first time?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 122.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [122.0, 137.0], "base_se_norm": [0.0469130277633667, 0.17563188076019287], "gt_se_raw": [123.0, 138.0], "gt_se_norm": [0.05549430847167969, 0.1842130422592163], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006651.pt"}
{"idx": 6653, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QjqS7jzjX34.mp4", "duration": 285.4184666666667, "prompt_text": "Please find the visual event described by the sentence 'When does Rogue lift the green tank and drop the soldiers out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Rogue lift the green tank and drop the soldiers out?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 27.0], "base_se_norm": [-0.8458403944969177, -0.8108041286468506], "gt_se_raw": [24.0, 28.0], "gt_se_norm": [-0.831825852394104, -0.8037968873977661], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006653.pt"}
{"idx": 6655, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QjqS7jzjX34.mp4", "duration": 285.4184666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where Storm summons a whirlwind to destroy a tank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 69.0 - 73.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where Storm summons a whirlwind to destroy a tank.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 69.0 - 73.0 seconds.<|im_end|>\n", "base_se_raw": [69.0, 73.0], "base_se_norm": [-0.516499400138855, -0.4884703755378723], "gt_se_raw": [69.0, 73.0], "gt_se_norm": [-0.516499400138855, -0.4884703755378723], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006655.pt"}
{"idx": 6657, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QjqS7jzjX34.mp4", "duration": 285.4184666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the scene where the woman with yellow hair uses her pink energy to trap the three generals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 277.0 - 283.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the scene where the woman with yellow hair uses her pink energy to trap the three generals.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 277.0 - 283.0 seconds.<|im_end|>\n", "base_se_raw": [277.0, 283.0], "base_se_norm": [0.9410097599029541, 0.9830532073974609], "gt_se_raw": [278.0, 282.0], "gt_se_norm": [0.9480170011520386, 0.9760459661483765], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006657.pt"}
{"idx": 6659, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/alLd1Bobkf8.mp4", "duration": 260.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'When are clues for a 'closed lid' state discussed, accompanied by a list and a picture of a child with their head on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 140.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are clues for a 'closed lid' state discussed, accompanied by a list and a picture of a child with their head on the table?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 140.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [140.0, 183.0], "base_se_norm": [0.07554411888122559, 0.4058898687362671], "gt_se_raw": [140.0, 184.0], "gt_se_norm": [0.07554411888122559, 0.4135723114013672], "start_token_span": [83, 88], "end_token_span": [90, 95], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006659.pt"}
{"idx": 6661, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/alLd1Bobkf8.mp4", "duration": 260.3333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video introduce the concept of the 'Box of Deafblindness'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video introduce the concept of the 'Box of Deafblindness'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 31.0], "base_se_norm": [-0.9462227821350098, -0.7618438005447388], "gt_se_raw": [8.0, 31.0], "gt_se_norm": [-0.9385403394699097, -0.7618438005447388], "start_token_span": [74, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006661.pt"}
{"idx": 6663, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iD2rhdFRehU.mp4", "duration": 291.16587499999997, "prompt_text": "Please find the visual event described by the sentence 'When does the shirtless man kiss the older man on the forehead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 57.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the shirtless man kiss the older man on the forehead?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 57.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 57.0], "base_se_norm": [-0.6222084760665894, -0.6084705591201782], "gt_se_raw": [55.0, 57.0], "gt_se_norm": [-0.6222084760665894, -0.6084705591201782], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006663.pt"}
{"idx": 6665, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iD2rhdFRehU.mp4", "duration": 291.16587499999997, "prompt_text": "Please find the visual event described by the sentence 'Show the moment the man and woman are dancing together in the living room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 137.0 - 157.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment the man and woman are dancing together in the living room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 137.0 - 157.0 seconds.<|im_end|>\n", "base_se_raw": [137.0, 157.0], "base_se_norm": [-0.058955609798431396, 0.0784231424331665], "gt_se_raw": [144.0, 150.0], "gt_se_norm": [-0.010873079299926758, 0.03034055233001709], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006665.pt"}
{"idx": 6667, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/iD2rhdFRehU.mp4", "duration": 291.16587499999997, "prompt_text": "Please find the visual event described by the sentence 'Show the man in a black suit levitating in the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 263.0 - 269.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man in a black suit levitating in the air.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 263.0 - 269.0 seconds.<|im_end|>\n", "base_se_raw": [263.0, 269.0], "base_se_norm": [0.8065304756164551, 0.8477441072463989], "gt_se_raw": [263.0, 265.0], "gt_se_norm": [0.8065304756164551, 0.8202682733535767], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006667.pt"}
{"idx": 6669, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsX3ATc3FbA.mp4", "duration": 252.50224999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where three men in pink outfits are dancing in front of the theater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where three men in pink outfits are dancing in front of the theater.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 48.0], "base_se_norm": [-0.6752504110336304, -0.6198053359985352], "gt_se_raw": [41.0, 46.0], "gt_se_norm": [-0.6752504110336304, -0.6356468200683594], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006669.pt"}
{"idx": 6671, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/XsX3ATc3FbA.mp4", "duration": 252.50224999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the group is jumping on and around a yellow sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the group is jumping on and around a yellow sofa.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 113.0], "base_se_norm": [-0.13664132356643677, -0.10495847463607788], "gt_se_raw": [109.0, 113.0], "gt_se_norm": [-0.13664132356643677, -0.10495847463607788], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006671.pt"}
{"idx": 6673, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgcISkB2duI.mp4", "duration": 282.6666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the boy is asked to put the letter 'H' into the box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 102.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the boy is asked to put the letter 'H' into the box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 102.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 102.0], "base_se_norm": [-0.32783013582229614, -0.27830183506011963], "gt_se_raw": [98.0, 102.0], "gt_se_norm": [-0.30660372972488403, -0.27830183506011963], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006673.pt"}
{"idx": 6675, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgcISkB2duI.mp4", "duration": 282.6666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show the time when the boy types 'baseball' on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 250.0 - 262.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the time when the boy types 'baseball' on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 250.0 - 262.0 seconds.<|im_end|>\n", "base_se_raw": [250.0, 262.0], "base_se_norm": [0.7688679695129395, 0.8537735939025879], "gt_se_raw": [202.0, 207.0], "gt_se_norm": [0.42924535274505615, 0.46462273597717285], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006675.pt"}
{"idx": 6677, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JgcISkB2duI.mp4", "duration": 282.6666666666667, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the scene where the boy is shown a toy snake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the scene where the boy is shown a toy snake.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 28.0], "base_se_norm": [-0.9292452931404114, -0.801886796951294], "gt_se_raw": [9.0, 16.0], "gt_se_norm": [-0.9363207817077637, -0.8867924213409424], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006677.pt"}
{"idx": 6679, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/--YU8YcWeUU.mp4", "duration": 228.09453333333335, "prompt_text": "Please find the visual event described by the sentence 'When can we see the woman in the purple jacket running on the track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the woman in the purple jacket running on the track?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.9210853576660156, -0.8947804570198059], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.929853618144989, -0.8947804570198059], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006679.pt"}
{"idx": 6681, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/--YU8YcWeUU.mp4", "duration": 228.09453333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the man demonstrates how to land in the middle of the foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 127.0 - 155.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the man demonstrates how to land in the middle of the foot.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 127.0 - 155.0 seconds.<|im_end|>\n", "base_se_raw": [127.0, 155.0], "base_se_norm": [0.11357343196868896, 0.35908567905426025], "gt_se_raw": [126.0, 136.0], "gt_se_norm": [0.1048051118850708, 0.19248807430267334], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006681.pt"}
{"idx": 6683, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FWlHYefkQms.mp4", "duration": 235.76793333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the brown jacket raise his hands as if to surrender?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the brown jacket raise his hands as if to surrender?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 12.0], "base_se_norm": [-0.9321366548538208, -0.8982049822807312], "gt_se_raw": [8.0, 12.0], "gt_se_norm": [-0.9321366548538208, -0.8982049822807312], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006683.pt"}
{"idx": 6685, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FWlHYefkQms.mp4", "duration": 235.76793333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does the young woman with a backpack look scared?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the young woman with a backpack look scared?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 142.0], "base_se_norm": [0.1451939344406128, 0.2045743465423584], "gt_se_raw": [134.0, 136.0], "gt_se_norm": [0.1367110013961792, 0.1536768674850464], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006685.pt"}
{"idx": 6687, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/FWlHYefkQms.mp4", "duration": 235.76793333333333, "prompt_text": "Please find the visual event described by the sentence 'When do they talk about the older man's son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 152.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do they talk about the older man's son?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 152.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 152.0], "base_se_norm": [0.1451939344406128, 0.2894035577774048], "gt_se_raw": [122.0, 143.0], "gt_se_norm": [0.034916043281555176, 0.213057279586792], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006687.pt"}
{"idx": 6689, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wu-gd9tLpmU.mp4", "duration": 221.98843333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the speaker give an auditory cue to help the girl count?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 42.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the speaker give an auditory cue to help the girl count?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 42.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 42.0], "base_se_norm": [-0.6756588220596313, -0.621601939201355], "gt_se_raw": [37.0, 43.0], "gt_se_norm": [-0.666649341583252, -0.6125924587249756], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006689.pt"}
{"idx": 6691, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Wu-gd9tLpmU.mp4", "duration": 221.98843333333335, "prompt_text": "Please find the visual event described by the sentence 'When is the girl's counting interrupted by her sister making a noise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the girl's counting interrupted by her sister making a noise?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 33.0], "base_se_norm": [-0.7567440867424011, -0.7026872038841248], "gt_se_raw": [30.0, 36.0], "gt_se_norm": [-0.7297156453132629, -0.6756588220596313], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006691.pt"}
{"idx": 6693, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/reZpm2zH3UQ.mp4", "duration": 282.53333333333336, "prompt_text": "Please find the visual event described by the sentence 'When does the user click the 'Subtitles' option in the YouTube studio menu?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 50.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the user click the 'Subtitles' option in the YouTube studio menu?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 50.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [50.0, 54.0], "base_se_norm": [-0.6460594534873962, -0.6177442073822021], "gt_se_raw": [49.0, 51.0], "gt_se_norm": [-0.653138279914856, -0.6389806270599365], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006693.pt"}
{"idx": 6695, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/reZpm2zH3UQ.mp4", "duration": 282.53333333333336, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the 'Add inline' and 'Add extended' options are being explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 55.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the 'Add inline' and 'Add extended' options are being explained.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 55.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [55.0, 88.0], "base_se_norm": [-0.6106654405593872, -0.37706464529037476], "gt_se_raw": [197.0, 211.0], "gt_se_norm": [0.39452576637268066, 0.4936290979385376], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006695.pt"}
{"idx": 6697, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GqSG43pgvZE.mp4", "duration": 223.89033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when the woman is eating a cake that looks like a gift box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the woman is eating a cake that looks like a gift box.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 128.0], "base_se_norm": [0.10768520832061768, 0.1434168815612793], "gt_se_raw": [124.0, 128.0], "gt_se_norm": [0.10768520832061768, 0.1434168815612793], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006697.pt"}
{"idx": 6699, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GqSG43pgvZE.mp4", "duration": 223.89033333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment a person cuts a green, square-shaped cake with a knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 156.0 - 159.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment a person cuts a green, square-shaped cake with a knife.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 156.0 - 159.0 seconds.<|im_end|>\n", "base_se_raw": [156.0, 159.0], "base_se_norm": [0.3935394287109375, 0.4203382730484009], "gt_se_raw": [157.0, 159.0], "gt_se_norm": [0.4024723768234253, 0.4203382730484009], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006699.pt"}
{"idx": 6701, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GqSG43pgvZE.mp4", "duration": 223.89033333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me when a woman is holding two macarons up to her eyes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 144.0 - 147.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when a woman is holding two macarons up to her eyes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 144.0 - 147.0 seconds.<|im_end|>\n", "base_se_raw": [144.0, 147.0], "base_se_norm": [0.286344051361084, 0.31314289569854736], "gt_se_raw": [144.0, 147.0], "gt_se_norm": [0.286344051361084, 0.31314289569854736], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006701.pt"}
{"idx": 6703, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5dMmKehyVUs.mp4", "duration": 298.56493333333333, "prompt_text": "Please find the visual event described by the sentence 'At what moment does the woman mention she is about to take a bath?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 43.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment does the woman mention she is about to take a bath?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 43.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 43.0], "base_se_norm": [-0.7320516109466553, -0.7119554281234741], "gt_se_raw": [75.0, 77.0], "gt_se_norm": [-0.49759674072265625, -0.4841993451118469], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006703.pt"}
{"idx": 6705, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5dMmKehyVUs.mp4", "duration": 298.56493333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip showing the baseball stadium from the audience's seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 114.0 - 173.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip showing the baseball stadium from the audience's seats.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 114.0 - 173.0 seconds.<|im_end|>\n", "base_se_raw": [114.0, 173.0], "base_se_norm": [-0.2363470196723938, 0.15887689590454102], "gt_se_raw": [127.0, 138.0], "gt_se_norm": [-0.14926379919052124, -0.07557797431945801], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006705.pt"}
{"idx": 6707, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QGGyYVsEGGo.mp4", "duration": 240.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the clip for catch number 8.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 57.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip for catch number 8.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 57.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [57.0, 72.0], "base_se_norm": [-0.5255271792411804, -0.40066593885421753], "gt_se_raw": [56.0, 72.0], "gt_se_norm": [-0.5338512659072876, -0.40066593885421753], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006707.pt"}
{"idx": 6709, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QGGyYVsEGGo.mp4", "duration": 240.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when David Tyree makes the catch against his helmet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 178.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when David Tyree makes the catch against his helmet.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 178.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 178.0], "base_se_norm": [0.4400665760040283, 0.481687068939209], "gt_se_raw": [173.0, 177.0], "gt_se_norm": [0.4400665760040283, 0.47336292266845703], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006709.pt"}
{"idx": 6711, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/QGGyYVsEGGo.mp4", "duration": 240.26666666666668, "prompt_text": "Please find the visual event described by the sentence 'When does Odell Beckham Jr. score a touchdown against the Cowboys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 219.0 - 233.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does Odell Beckham Jr. score a touchdown against the Cowboys?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 219.0 - 233.0 seconds.<|im_end|>\n", "base_se_raw": [219.0, 233.0], "base_se_norm": [0.8229745626449585, 0.9395116567611694], "gt_se_raw": [223.0, 226.0], "gt_se_norm": [0.8562707901000977, 0.881243109703064], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006711.pt"}
{"idx": 6713, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U11DgbefmQQ.mp4", "duration": 310.0, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the mantis shrimp drags the crab it just killed into its den under the rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 45.0 - 54.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the mantis shrimp drags the crab it just killed into its den under the rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 45.0 - 54.0 seconds.<|im_end|>\n", "base_se_raw": [45.0, 54.0], "base_se_norm": [-0.7096774578094482, -0.6516128778457642], "gt_se_raw": [119.0, 126.0], "gt_se_norm": [-0.23225808143615723, -0.18709677457809448], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006713.pt"}
{"idx": 6715, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/U11DgbefmQQ.mp4", "duration": 310.0, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the mantis shrimp attacks a crab near a white pipe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 147.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the mantis shrimp attacks a crab near a white pipe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 147.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [147.0, 158.0], "base_se_norm": [-0.051612913608551025, 0.019354820251464844], "gt_se_raw": [138.0, 146.0], "gt_se_norm": [-0.10967743396759033, -0.05806452035903931], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006715.pt"}
{"idx": 6717, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JaU55dchC-g.mp4", "duration": 272.272, "prompt_text": "Please find the visual event described by the sentence 'At what moment is Andrew Martin being introduced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what moment is Andrew Martin being introduced?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 70.0], "base_se_norm": [-0.5372275114059448, -0.4858083128929138], "gt_se_raw": [63.0, 69.0], "gt_se_norm": [-0.5372275114059448, -0.4931539297103882], "start_token_span": [63, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006717.pt"}
{"idx": 6719, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JaU55dchC-g.mp4", "duration": 272.272, "prompt_text": "Please find the visual event described by the sentence 'When are the New Century Scholars honored?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 219.0 - 259.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the New Century Scholars honored?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 219.0 - 259.0 seconds.<|im_end|>\n", "base_se_raw": [219.0, 259.0], "base_se_norm": [0.6086853742599487, 0.9025092124938965], "gt_se_raw": [219.0, 247.0], "gt_se_norm": [0.6086853742599487, 0.8143620491027832], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006719.pt"}
{"idx": 6721, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JaU55dchC-g.mp4", "duration": 272.272, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range for the introduction of Jennifer Gomez Villegas from Hopkinsville Community College.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 113.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range for the introduction of Jennifer Gomez Villegas from Hopkinsville Community College.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 113.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 113.0], "base_se_norm": [-0.21402126550674438, -0.16994768381118774], "gt_se_raw": [107.0, 114.0], "gt_se_norm": [-0.21402126550674438, -0.16260212659835815], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006721.pt"}
{"idx": 6723, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/14GZruL1hrE.mp4", "duration": 299.2572916666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is the word 'daughters' shown in the will?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 104.0 - 109.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the word 'daughters' shown in the will?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 104.0 - 109.0 seconds.<|im_end|>\n", "base_se_raw": [104.0, 109.0], "base_se_norm": [-0.3049459457397461, -0.2715298533439636], "gt_se_raw": [105.0, 107.0], "gt_se_norm": [-0.29826271533966064, -0.2848963141441345], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006723.pt"}
{"idx": 6725, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/14GZruL1hrE.mp4", "duration": 299.2572916666666, "prompt_text": "Please find the visual event described by the sentence 'Show the clip where the blonde girl points a gun at the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 216.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the clip where the blonde girl points a gun at the man.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 216.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [216.0, 222.0], "base_se_norm": [0.44357383251190186, 0.483673095703125], "gt_se_raw": [216.0, 219.0], "gt_se_norm": [0.44357383251190186, 0.4636235237121582], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006725.pt"}
{"idx": 6727, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4eYAVZkCjvY.mp4", "duration": 290.55693333333335, "prompt_text": "Please find the visual event described by the sentence 'Show me when the child dressed as Steve Harvey is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the child dressed as Steve Harvey is introduced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 24.0], "base_se_norm": [-0.9242833256721497, -0.8348000049591064], "gt_se_raw": [11.0, 20.0], "gt_se_norm": [-0.9242833256721497, -0.862333357334137], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006727.pt"}
{"idx": 6729, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4eYAVZkCjvY.mp4", "duration": 290.55693333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point does the child in the Shutterfly check costume appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 121.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the child in the Shutterfly check costume appear?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 121.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [121.0, 128.0], "base_se_norm": [-0.16711682081222534, -0.11893349885940552], "gt_se_raw": [122.0, 128.0], "gt_se_norm": [-0.16023343801498413, -0.11893349885940552], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006729.pt"}
{"idx": 6731, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4eYAVZkCjvY.mp4", "duration": 290.55693333333335, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the commercial for milk featuring a UFO is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 226.0 - 242.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the commercial for milk featuring a UFO is shown.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 226.0 - 242.0 seconds.<|im_end|>\n", "base_se_raw": [226.0, 242.0], "base_se_norm": [0.5556330680847168, 0.6657663583755493], "gt_se_raw": [225.0, 242.0], "gt_se_norm": [0.5487498044967651, 0.6657663583755493], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006731.pt"}
{"idx": 6733, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XzP3jOJn5c.mp4", "duration": 223.24, "prompt_text": "Please find the visual event described by the sentence 'Show the scene that demonstrates the parasol effect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 94.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene that demonstrates the parasol effect.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 94.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 94.0], "base_se_norm": [-0.23848772048950195, -0.15785706043243408], "gt_se_raw": [87.0, 94.0], "gt_se_norm": [-0.22056978940963745, -0.15785706043243408], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006733.pt"}
{"idx": 6735, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/9XzP3jOJn5c.mp4", "duration": 223.24, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the different origins of the aerosol are illustrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 36.0 - 53.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the different origins of the aerosol are illustrated.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 36.0 - 53.0 seconds.<|im_end|>\n", "base_se_raw": [36.0, 53.0], "base_se_norm": [-0.6774771213531494, -0.5251747369766235], "gt_se_raw": [36.0, 45.0], "gt_se_norm": [-0.6774771213531494, -0.5968464612960815], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006735.pt"}
{"idx": 6737, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IW40rIEXKXs.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'When does the man say 'All this economic crap is now connected'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 10.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man say 'All this economic crap is now connected'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 10.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [10.0, 13.0], "base_se_norm": [-0.9099999070167542, -0.8829998970031738], "gt_se_raw": [32.0, 35.0], "gt_se_norm": [-0.7119997143745422, -0.6849997043609619], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006737.pt"}
{"idx": 6739, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IW40rIEXKXs.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the speaker talks about a Chinese stock market burst.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 41.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the speaker talks about a Chinese stock market burst.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 41.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [41.0, 59.0], "base_se_norm": [-0.6309996247291565, -0.468999445438385], "gt_se_raw": [53.0, 56.0], "gt_se_norm": [-0.5229995250701904, -0.4959995150566101], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006739.pt"}
{"idx": 6741, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/IW40rIEXKXs.mp4", "duration": 222.222, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where he mentions socialist and capitalist leaning folks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 40.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where he mentions socialist and capitalist leaning folks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 40.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [40.0, 50.0], "base_se_norm": [-0.6399996280670166, -0.5499995946884155], "gt_se_raw": [122.0, 125.0], "gt_se_norm": [0.09800112247467041, 0.12500107288360596], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006741.pt"}
{"idx": 6743, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bzua8Zvlppo.mp4", "duration": 209.0088, "prompt_text": "Please find the visual event described by the sentence 'At what point does a speaker discuss how maintaining the organization itself becomes the primary work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 37.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a speaker discuss how maintaining the organization itself becomes the primary work?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 37.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 37.0], "base_se_norm": [-0.8947412967681885, -0.6459479331970215], "gt_se_raw": [42.0, 47.0], "gt_se_norm": [-0.5981030464172363, -0.5502581596374512], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006743.pt"}
{"idx": 6745, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bzua8Zvlppo.mp4", "duration": 209.0088, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a woman with curly hair talks about losing the 'bigger picture'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 116.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a woman with curly hair talks about losing the 'bigger picture'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 116.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 116.0], "base_se_norm": [-0.2631889581680298, 0.1100010871887207], "gt_se_raw": [77.0, 81.0], "gt_se_norm": [-0.2631889581680298, -0.22491306066513062], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006745.pt"}
{"idx": 6747, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a5V6gdu5ih8.mp4", "duration": 213.76, "prompt_text": "Please find the visual event described by the sentence 'When can we see the lion jumping over the stream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 12.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the lion jumping over the stream?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 12.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 12.0], "base_se_norm": [-0.9157934188842773, -0.8877245187759399], "gt_se_raw": [9.0, 11.0], "gt_se_norm": [-0.9157934188842773, -0.8970808386802673], "start_token_span": [65, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006747.pt"}
{"idx": 6749, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a5V6gdu5ih8.mp4", "duration": 213.76, "prompt_text": "Please find the visual event described by the sentence 'Show the wide shot where the lion is in the middle of a field with many hyenas approaching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 33.0 - 44.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the wide shot where the lion is in the middle of a field with many hyenas approaching.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 33.0 - 44.0 seconds.<|im_end|>\n", "base_se_raw": [33.0, 44.0], "base_se_norm": [-0.691242516040802, -0.5883233547210693], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.7005988359451294, -0.6631736755371094], "start_token_span": [74, 78], "end_token_span": [80, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006749.pt"}
{"idx": 6751, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/a5V6gdu5ih8.mp4", "duration": 213.76, "prompt_text": "Please find the visual event described by the sentence 'When are the two lions reunited and rubbing their heads together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 192.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two lions reunited and rubbing their heads together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 192.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 192.0], "base_se_norm": [0.7028443813323975, 0.7964072227478027], "gt_se_raw": [182.0, 192.0], "gt_se_norm": [0.7028443813323975, 0.7964072227478027], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006751.pt"}
{"idx": 6753, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tndpRiDf9Z0.mp4", "duration": 215.840625, "prompt_text": "Please find the visual event described by the sentence 'When does a woman in a floral dress walk in from the left side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does a woman in a floral dress walk in from the left side of the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 40.0], "base_se_norm": [-0.7034848928451538, -0.6293561458587646], "gt_se_raw": [33.0, 37.0], "gt_se_norm": [-0.6942188143730164, -0.6571544408798218], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006753.pt"}
{"idx": 6755, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/tndpRiDf9Z0.mp4", "duration": 215.840625, "prompt_text": "Please find the visual event described by the sentence 'Locate the part where the woman argues that her work is for the greater good of the universe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the part where the woman argues that her work is for the greater good of the universe.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 118.0], "base_se_norm": [-0.12898695468902588, 0.09339940547943115], "gt_se_raw": [100.0, 105.0], "gt_se_norm": [-0.07339036464691162, -0.027059853076934814], "start_token_span": [74, 78], "end_token_span": [80, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006755.pt"}
{"idx": 6757, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t1PGWO2Lvmw.mp4", "duration": 203.66179166666666, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the crepe batter being whisked until smooth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 39.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the crepe batter being whisked until smooth?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 39.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 39.0], "base_se_norm": [-0.6562929153442383, -0.6170121431350708], "gt_se_raw": [36.0, 40.0], "gt_se_norm": [-0.646472692489624, -0.6071919202804565], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006757.pt"}
{"idx": 6759, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t1PGWO2Lvmw.mp4", "duration": 203.66179166666666, "prompt_text": "Please find the visual event described by the sentence 'At what point is a crepe being flipped in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a crepe being flipped in the pan?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 74.0], "base_se_norm": [-0.2929454445838928, -0.2733049988746643], "gt_se_raw": [72.0, 74.0], "gt_se_norm": [-0.2929454445838928, -0.2733049988746643], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006759.pt"}
{"idx": 6761, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/t1PGWO2Lvmw.mp4", "duration": 203.66179166666666, "prompt_text": "Please find the visual event described by the sentence 'When are all the rolled crepes assembled together to form the large rose cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 118.0 - 137.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are all the rolled crepes assembled together to form the large rose cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 118.0 - 137.0 seconds.<|im_end|>\n", "base_se_raw": [118.0, 137.0], "base_se_norm": [0.1587839126586914, 0.34536778926849365], "gt_se_raw": [118.0, 137.0], "gt_se_norm": [0.1587839126586914, 0.34536778926849365], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006761.pt"}
{"idx": 6763, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7FgIp5QrZik.mp4", "duration": 301.926625, "prompt_text": "Please find the visual event described by the sentence 'What is the time when the book shows Leon turning purple on the yellow sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time when the book shows Leon turning purple on the yellow sand?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 62.0], "base_se_norm": [-0.6555454730987549, -0.5893042087554932], "gt_se_raw": [53.0, 61.0], "gt_se_norm": [-0.6489213705062866, -0.5959283113479614], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006763.pt"}
{"idx": 6765, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/7FgIp5QrZik.mp4", "duration": 301.926625, "prompt_text": "Please find the visual event described by the sentence 'At what point is Leon, colored red, hiding under a leaf, unable to camouflage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is Leon, colored red, hiding under a leaf, unable to camouflage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 99.0], "base_se_norm": [-0.4303252100944519, -0.3442115783691406], "gt_se_raw": [89.0, 99.0], "gt_se_norm": [-0.41045278310775757, -0.3442115783691406], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006765.pt"}
{"idx": 6767, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwXx63AoSwU.mp4", "duration": 277.4583333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the Evil Queen look into the magic mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 28.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the Evil Queen look into the magic mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 28.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [28.0, 36.0], "base_se_norm": [-0.798167884349823, -0.7405015826225281], "gt_se_raw": [28.0, 32.0], "gt_se_norm": [-0.798167884349823, -0.7693347334861755], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006767.pt"}
{"idx": 6769, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwXx63AoSwU.mp4", "duration": 277.4583333333333, "prompt_text": "Please find the visual event described by the sentence 'Find the moment Jafar is holding the genie lamp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 75.0 - 82.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment Jafar is holding the genie lamp.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 75.0 - 82.0 seconds.<|im_end|>\n", "base_se_raw": [75.0, 82.0], "base_se_norm": [-0.45937830209732056, -0.4089202880859375], "gt_se_raw": [90.0, 93.0], "gt_se_norm": [-0.3512539863586426, -0.329629123210907], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006769.pt"}
{"idx": 6771, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/WwXx63AoSwU.mp4", "duration": 277.4583333333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do all the villains cast a spell on the cauldron together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 181.0 - 221.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do all the villains cast a spell on the cauldron together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 181.0 - 221.0 seconds.<|im_end|>\n", "base_se_raw": [181.0, 221.0], "base_se_norm": [0.3047003746032715, 0.5930318832397461], "gt_se_raw": [193.0, 198.0], "gt_se_norm": [0.39119982719421387, 0.42724132537841797], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006771.pt"}
{"idx": 6773, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y9PV2rfjJo4.mp4", "duration": 236.5363, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker describe the young man's excited reaction to the idea of talking to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker describe the young man's excited reaction to the idea of talking to the girl?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 114.0], "base_se_norm": [-0.15446382761001587, -0.036088764667510986], "gt_se_raw": [86.0, 89.0], "gt_se_norm": [-0.27283889055252075, -0.2474728226661682], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006773.pt"}
{"idx": 6775, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/y9PV2rfjJo4.mp4", "duration": 236.5363, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the speaker describes how she advised the young man to sit beside the girls instead of standing over them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the speaker describes how she advised the young man to sit beside the girls instead of standing over them.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 114.0], "base_se_norm": [-0.15446382761001587, -0.036088764667510986], "gt_se_raw": [159.0, 175.0], "gt_se_norm": [0.344402551651001, 0.47968828678131104], "start_token_span": [78, 83], "end_token_span": [85, 90], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006775.pt"}
{"idx": 6777, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jofNR_WkoCE.mp4", "duration": 224.8, "prompt_text": "Please find the visual event described by the sentence 'Show the scene where people dressed as animals are having a party on the balcony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the scene where people dressed as animals are having a party on the balcony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 13.0], "base_se_norm": [-0.9644128084182739, -0.8843416571617126], "gt_se_raw": [1.0, 4.0], "gt_se_norm": [-0.9911032319068909, -0.9644128084182739], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006777.pt"}
{"idx": 6779, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jofNR_WkoCE.mp4", "duration": 224.8, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the old man is reading a book to the child in the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 68.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the old man is reading a book to the child in the forest.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 68.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 68.0], "base_se_norm": [-0.4395017623901367, -0.3950178027153015], "gt_se_raw": [63.0, 65.0], "gt_se_norm": [-0.4395017623901367, -0.4217081665992737], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006779.pt"}
{"idx": 6781, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/jofNR_WkoCE.mp4", "duration": 224.8, "prompt_text": "Please find the visual event described by the sentence 'When are two men in fox costumes shown levitating in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 154.0 - 158.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are two men in fox costumes shown levitating in the air?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 154.0 - 158.0 seconds.<|im_end|>\n", "base_se_raw": [154.0, 158.0], "base_se_norm": [0.37010669708251953, 0.4056938886642456], "gt_se_raw": [155.0, 157.0], "gt_se_norm": [0.37900352478027344, 0.39679718017578125], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006781.pt"}
{"idx": 6783, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLjHyP683QU.mp4", "duration": 205.04, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a yellow crab is wrapping gifts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a yellow crab is wrapping gifts.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 88.0], "base_se_norm": [-0.21966445446014404, -0.1416308879852295], "gt_se_raw": [80.0, 85.0], "gt_se_norm": [-0.21966445446014404, -0.17089343070983887], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006783.pt"}
{"idx": 6785, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/aLjHyP683QU.mp4", "duration": 205.04, "prompt_text": "Please find the visual event described by the sentence 'At what point do the turtle and jellyfish hang up the decorations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 119.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do the turtle and jellyfish hang up the decorations?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 119.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 119.0], "base_se_norm": [0.08271563053131104, 0.16074919700622559], "gt_se_raw": [112.0, 118.0], "gt_se_norm": [0.09246981143951416, 0.15099501609802246], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006785.pt"}
{"idx": 6787, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zRgZ3sWvnqs.mp4", "duration": 269.8112083333333, "prompt_text": "Please find the visual event described by the sentence 'When are the two speakers sitting outdoors with a view of the water and a city?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 23.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two speakers sitting outdoors with a view of the water and a city?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 23.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 23.0], "base_se_norm": [-1.0, -0.8295104503631592], "gt_se_raw": [0.0, 23.0], "gt_se_norm": [-1.0, -0.8295104503631592], "start_token_span": [71, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006787.pt"}
{"idx": 6789, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zRgZ3sWvnqs.mp4", "duration": 269.8112083333333, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the speaker with the red shirt is explaining how to search for a book using metadata.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 143.0 - 154.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the speaker with the red shirt is explaining how to search for a book using metadata.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 143.0 - 154.0 seconds.<|im_end|>\n", "base_se_raw": [143.0, 154.0], "base_se_norm": [0.06000041961669922, 0.1415388584136963], "gt_se_raw": [143.0, 154.0], "gt_se_norm": [0.06000041961669922, 0.1415388584136963], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006789.pt"}
{"idx": 6791, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zRgZ3sWvnqs.mp4", "duration": 269.8112083333333, "prompt_text": "Please find the visual event described by the sentence 'Find the clip where the two men are sitting in green chairs and using tablets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 210.0 - 227.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the clip where the two men are sitting in green chairs and using tablets.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 210.0 - 227.0 seconds.<|im_end|>\n", "base_se_raw": [210.0, 227.0], "base_se_norm": [0.5566439628601074, 0.6826579570770264], "gt_se_raw": [210.0, 218.0], "gt_se_norm": [0.5566439628601074, 0.6159446239471436], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006791.pt"}
{"idx": 6793, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hPLgfGX1I5Y.mp4", "duration": 206.8733333333333, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an x-ray of a dinosaur's skeleton is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 52.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an x-ray of a dinosaur's skeleton is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 52.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [52.0, 59.0], "base_se_norm": [-0.4972769021987915, -0.4296026825904846], "gt_se_raw": [52.0, 56.0], "gt_se_norm": [-0.4972769021987915, -0.45860594511032104], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006793.pt"}
{"idx": 6795, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/hPLgfGX1I5Y.mp4", "duration": 206.8733333333333, "prompt_text": "Please find the visual event described by the sentence 'Where can I find the animation of a peacock displaying its magnificent tail feathers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 149.0 - 153.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where can I find the animation of a peacock displaying its magnificent tail feathers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 149.0 - 153.0 seconds.<|im_end|>\n", "base_se_raw": [149.0, 153.0], "base_se_norm": [0.44049501419067383, 0.4791659116744995], "gt_se_raw": [149.0, 152.0], "gt_se_norm": [0.44049501419067383, 0.46949827671051025], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006795.pt"}
{"idx": 6797, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c346ms-PaeE.mp4", "duration": 207.33333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see a man in a black coat holding a bow and arrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 92.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a man in a black coat holding a bow and arrow?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 92.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 92.0], "base_se_norm": [-0.17041796445846558, -0.11254018545150757], "gt_se_raw": [145.0, 146.0], "gt_se_norm": [0.3987138271331787, 0.4083601236343384], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006797.pt"}
{"idx": 6799, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c346ms-PaeE.mp4", "duration": 207.33333333333334, "prompt_text": "Please find the visual event described by the sentence 'When does the person in blue and black plaid kill the man in the blue shirt and black pants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person in blue and black plaid kill the man in the blue shirt and black pants?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 168.0], "base_se_norm": [0.5916398763656616, 0.6205787658691406], "gt_se_raw": [166.0, 168.0], "gt_se_norm": [0.6012861728668213, 0.6205787658691406], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006799.pt"}
{"idx": 6801, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/c346ms-PaeE.mp4", "duration": 207.33333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the man in the black shirt and yellow T-shirt killed with a knife?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man in the black shirt and yellow T-shirt killed with a knife?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 182.0], "base_se_norm": [0.7073955535888672, 0.7556270360946655], "gt_se_raw": [178.0, 180.0], "gt_se_norm": [0.7170418500900269, 0.7363344430923462], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006801.pt"}
{"idx": 6803, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1fB0BKm9dTE.mp4", "duration": 311.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the tiger sniff the ground near the tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 31.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the tiger sniff the ground near the tree?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 31.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 31.0], "base_se_norm": [-0.8457252979278564, -0.8007284998893738], "gt_se_raw": [25.0, 28.0], "gt_se_norm": [-0.8392971754074097, -0.8200128674507141], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006803.pt"}
{"idx": 6805, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1fB0BKm9dTE.mp4", "duration": 311.1333333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time is a lion seen resting on a large pipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 120.0 - 139.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time is a lion seen resting on a large pipe?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 120.0 - 139.0 seconds.<|im_end|>\n", "base_se_raw": [120.0, 139.0], "base_se_norm": [-0.228626549243927, -0.10649240016937256], "gt_se_raw": [120.0, 138.0], "gt_se_norm": [-0.228626549243927, -0.11292052268981934], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006805.pt"}
{"idx": 6807, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TSVj63OE9qM.mp4", "duration": 212.04, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where a queen in a black dress with a gold crown performs by herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 50.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where a queen in a black dress with a gold crown performs by herself.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 50.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 50.0], "base_se_norm": [-0.8207885026931763, -0.5283908843994141], "gt_se_raw": [21.0, 51.0], "gt_se_norm": [-0.8019241690635681, -0.5189586877822876], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006807.pt"}
{"idx": 6809, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TSVj63OE9qM.mp4", "duration": 212.04, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the moment a woman in a fur coat and two-toned hair is seen singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the moment a woman in a fur coat and two-toned hair is seen singing.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 122.0], "base_se_norm": [-0.21712881326675415, 0.150726318359375], "gt_se_raw": [83.0, 122.0], "gt_se_norm": [-0.21712881326675415, 0.150726318359375], "start_token_span": [73, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006809.pt"}
{"idx": 6811, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/TSVj63OE9qM.mp4", "duration": 212.04, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range when four characters are displayed simultaneously on screen, all singing the final chorus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 164.0 - 189.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range when four characters are displayed simultaneously on screen, all singing the final chorus.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 164.0 - 189.0 seconds.<|im_end|>\n", "base_se_raw": [164.0, 189.0], "base_se_norm": [0.5468779802322388, 0.7826825380325317], "gt_se_raw": [164.0, 188.0], "gt_se_norm": [0.5468779802322388, 0.7732503414154053], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006811.pt"}
{"idx": 6813, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OstpJCHpTis.mp4", "duration": 280.0464333333333, "prompt_text": "Please find the visual event described by the sentence 'Please specify the time range when the old-fashioned syringes are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 18.0 - 20.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Please specify the time range when the old-fashioned syringes are displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 18.0 - 20.0 seconds.<|im_end|>\n", "base_se_raw": [18.0, 20.0], "base_se_norm": [-0.871449887752533, -0.8571665287017822], "gt_se_raw": [18.0, 20.0], "gt_se_norm": [-0.871449887752533, -0.8571665287017822], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006813.pt"}
{"idx": 6815, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OstpJCHpTis.mp4", "duration": 280.0464333333333, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when the name 'Edward Jenner' is prominently displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 210.0 - 212.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when the name 'Edward Jenner' is prominently displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 210.0 - 212.0 seconds.<|im_end|>\n", "base_se_raw": [210.0, 212.0], "base_se_norm": [0.4997512102127075, 0.5140345096588135], "gt_se_raw": [209.0, 213.0], "gt_se_norm": [0.49260950088500977, 0.5211762189865112], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006815.pt"}
{"idx": 6817, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ovyEMlNZlTE.mp4", "duration": 242.367125, "prompt_text": "Please find the visual event described by the sentence 'When are the two people hanging the painting on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the two people hanging the painting on the wall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 16.0], "base_se_norm": [-0.9009766578674316, -0.8679689168930054], "gt_se_raw": [0.0, 7.0], "gt_se_norm": [-1.0, -0.9422363638877869], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006817.pt"}
{"idx": 6819, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ovyEMlNZlTE.mp4", "duration": 242.367125, "prompt_text": "Please find the visual event described by the sentence 'Show the part where three people are sitting and having a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 122.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where three people are sitting and having a discussion.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 122.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 122.0], "base_se_norm": [-0.8019533157348633, 0.006737232208251953], "gt_se_raw": [56.0, 66.0], "gt_se_norm": [-0.537891149520874, -0.4553716778755188], "start_token_span": [67, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006819.pt"}
{"idx": 6821, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ovyEMlNZlTE.mp4", "duration": 242.367125, "prompt_text": "Please find the visual event described by the sentence 'Identify the time when the man in the purple jacket stands up from his chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 177.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time when the man in the purple jacket stands up from his chair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 177.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [177.0, 183.0], "base_se_norm": [0.46059417724609375, 0.5101057291030884], "gt_se_raw": [177.0, 182.0], "gt_se_norm": [0.46059417724609375, 0.5018538236618042], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006821.pt"}
{"idx": 6823, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZKVBUjaxOs.mp4", "duration": 251.5513, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when someone is adding seasonings to the white bean mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 79.0 - 87.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when someone is adding seasonings to the white bean mixture.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 79.0 - 87.0 seconds.<|im_end|>\n", "base_se_raw": [79.0, 87.0], "base_se_norm": [-0.37189751863479614, -0.3082922101020813], "gt_se_raw": [82.0, 87.0], "gt_se_norm": [-0.3480455279350281, -0.3082922101020813], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006823.pt"}
{"idx": 6825, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6ZKVBUjaxOs.mp4", "duration": 251.5513, "prompt_text": "Please find the visual event described by the sentence 'Find the part where the flank steak is being sliced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 215.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the part where the flank steak is being sliced.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 215.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [215.0, 222.0], "base_se_norm": [0.7093929052352905, 0.765047550201416], "gt_se_raw": [216.0, 222.0], "gt_se_norm": [0.7173435688018799, 0.765047550201416], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006825.pt"}
{"idx": 6827, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4ztYjFxwmI.mp4", "duration": 280.34673333333336, "prompt_text": "Please find the visual event described by the sentence 'When does the person pour ice cubes into the large bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 25.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the person pour ice cubes into the large bag?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 25.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [25.0, 38.0], "base_se_norm": [-0.821649432182312, -0.7289071083068848], "gt_se_raw": [25.0, 38.0], "gt_se_norm": [-0.821649432182312, -0.7289071083068848], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006827.pt"}
{"idx": 6829, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4ztYjFxwmI.mp4", "duration": 280.34673333333336, "prompt_text": "Please find the visual event described by the sentence 'At what point are the half-and-half milk and vanilla extract mixed together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 117.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the half-and-half milk and vanilla extract mixed together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 117.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 117.0], "base_se_norm": [-0.3222678303718567, -0.1653193235397339], "gt_se_raw": [94.0, 113.0], "gt_se_norm": [-0.3294018507003784, -0.19385546445846558], "start_token_span": [69, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006829.pt"}
{"idx": 6831, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/N4ztYjFxwmI.mp4", "duration": 280.34673333333336, "prompt_text": "Please find the visual event described by the sentence 'Find the time range where the finished ice cream is being scooped into a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 244.0 - 252.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range where the finished ice cream is being scooped into a bowl.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 244.0 - 252.0 seconds.<|im_end|>\n", "base_se_raw": [244.0, 252.0], "base_se_norm": [0.7407015562057495, 0.7977737188339233], "gt_se_raw": [243.0, 252.0], "gt_se_norm": [0.733567476272583, 0.7977737188339233], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006831.pt"}
{"idx": 6833, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-GRl8urBK4c.mp4", "duration": 302.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the red wool is being poked with a needle to form a round shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 54.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the red wool is being poked with a needle to form a round shape.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 54.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [54.0, 112.0], "base_se_norm": [-0.6425813436508179, -0.25868725776672363], "gt_se_raw": [57.0, 62.0], "gt_se_norm": [-0.6227247714996338, -0.5896304249763489], "start_token_span": [75, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006833.pt"}
{"idx": 6835, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/-GRl8urBK4c.mp4", "duration": 302.1666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the white wool being rolled into a long log for the stalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 173.0 - 187.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the white wool being rolled into a long log for the stalk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 173.0 - 187.0 seconds.<|im_end|>\n", "base_se_raw": [173.0, 187.0], "base_se_norm": [0.14506351947784424, 0.23772752285003662], "gt_se_raw": [172.0, 186.0], "gt_se_norm": [0.1384446620941162, 0.2311086654663086], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006835.pt"}
{"idx": 6837, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PB9JxYt2ls.mp4", "duration": 239.3391, "prompt_text": "Please find the visual event described by the sentence 'When are the materials needed for the craft shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 16.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the materials needed for the craft shown?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 16.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 16.0], "base_se_norm": [-0.9415056109428406, -0.866298496723175], "gt_se_raw": [7.0, 15.0], "gt_se_norm": [-0.9415056109428406, -0.8746548295021057], "start_token_span": [64, 67], "end_token_span": [69, 73], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006837.pt"}
{"idx": 6839, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PB9JxYt2ls.mp4", "duration": 239.3391, "prompt_text": "Please find the visual event described by the sentence 'Show the step where the sides are pushed inwards and flattened to form a square base.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 80.0 - 105.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the step where the sides are pushed inwards and flattened to form a square base.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 80.0 - 105.0 seconds.<|im_end|>\n", "base_se_raw": [80.0, 105.0], "base_se_norm": [-0.33149242401123047, -0.12258380651473999], "gt_se_raw": [81.0, 104.0], "gt_se_norm": [-0.3231360912322998, -0.13094013929367065], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006839.pt"}
{"idx": 6841, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/5PB9JxYt2ls.mp4", "duration": 239.3391, "prompt_text": "Please find the visual event described by the sentence 'Locate the segment where the face is being drawn on the completed pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 223.0 - 231.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the segment where the face is being drawn on the completed pumpkin.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 223.0 - 231.0 seconds.<|im_end|>\n", "base_se_raw": [223.0, 231.0], "base_se_norm": [0.8634648323059082, 0.9303156137466431], "gt_se_raw": [222.0, 231.0], "gt_se_norm": [0.8551084995269775, 0.9303156137466431], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006841.pt"}
{"idx": 6843, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5H8zBb3iao.mp4", "duration": 226.97674999999998, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when the men are dancing in colorful patchwork clothes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 26.0 - 30.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when the men are dancing in colorful patchwork clothes.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 26.0 - 30.0 seconds.<|im_end|>\n", "base_se_raw": [26.0, 30.0], "base_se_norm": [-0.770901620388031, -0.7356557250022888], "gt_se_raw": [27.0, 30.0], "gt_se_norm": [-0.7620901465415955, -0.7356557250022888], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006843.pt"}
{"idx": 6845, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/A5H8zBb3iao.mp4", "duration": 226.97674999999998, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where 'WARNING!!' is displayed on the TV screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where 'WARNING!!' is displayed on the TV screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 112.0], "base_se_norm": [-0.030737698078155518, -0.013114750385284424], "gt_se_raw": [110.0, 111.0], "gt_se_norm": [-0.030737698078155518, -0.02192622423171997], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006845.pt"}
{"idx": 6847, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Yrcr9QOnqB4.mp4", "duration": 313.4798333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the man getting up from his chair and walking towards the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 5.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the man getting up from his chair and walking towards the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 5.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [5.0, 13.0], "base_se_norm": [-0.9681000113487244, -0.9170600771903992], "gt_se_raw": [5.0, 10.0], "gt_se_norm": [-0.9681000113487244, -0.9362000226974487], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006847.pt"}
{"idx": 6849, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Yrcr9QOnqB4.mp4", "duration": 313.4798333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the dog bark while sitting in its basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 86.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the dog bark while sitting in its basket?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 86.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [86.0, 88.0], "base_se_norm": [-0.4513203501701355, -0.4385603666305542], "gt_se_raw": [86.0, 88.0], "gt_se_norm": [-0.4513203501701355, -0.4385603666305542], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006849.pt"}
{"idx": 6851, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Yrcr9QOnqB4.mp4", "duration": 313.4798333333333, "prompt_text": "Please find the visual event described by the sentence 'Show the police officer running towards the house filled with smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 243.0 - 259.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the police officer running towards the house filled with smoke.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 243.0 - 259.0 seconds.<|im_end|>\n", "base_se_raw": [243.0, 259.0], "base_se_norm": [0.5503389835357666, 0.6524189710617065], "gt_se_raw": [243.0, 257.0], "gt_se_norm": [0.5503389835357666, 0.6396589279174805], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006851.pt"}
{"idx": 6853, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ml9EdqgtVfU.mp4", "duration": 299.23333333333335, "prompt_text": "Please find the visual event described by the sentence 'Where in the video does it demonstrate how to add alternative text to an image?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 51.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Where in the video does it demonstrate how to add alternative text to an image?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 51.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [51.0, 65.0], "base_se_norm": [-0.6591289043426514, -0.5655564069747925], "gt_se_raw": [32.0, 40.0], "gt_se_norm": [-0.7861200571060181, -0.732650101184845], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006853.pt"}
{"idx": 6855, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ml9EdqgtVfU.mp4", "duration": 299.23333333333335, "prompt_text": "Please find the visual event described by the sentence 'At what point is a YouTube video URL copied and pasted to be embedded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 178.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is a YouTube video URL copied and pasted to be embedded?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 178.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [178.0, 193.0], "base_se_norm": [0.18970704078674316, 0.2899632453918457], "gt_se_raw": [179.0, 189.0], "gt_se_norm": [0.19639074802398682, 0.263228178024292], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006855.pt"}
{"idx": 6857, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-zXT5bIBM0.mp4", "duration": 201.73486666666668, "prompt_text": "Please find the visual event described by the sentence 'When does the little girl first look under the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the little girl first look under the door?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 21.0], "base_se_norm": [-0.8413759469985962, -0.7918059229850769], "gt_se_raw": [17.0, 20.0], "gt_se_norm": [-0.8314619660377502, -0.8017199635505676], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006857.pt"}
{"idx": 6859, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-zXT5bIBM0.mp4", "duration": 201.73486666666668, "prompt_text": "Please find the visual event described by the sentence 'At what point does the girl ride a bicycle in the hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 58.0 - 62.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the girl ride a bicycle in the hall?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 58.0 - 62.0 seconds.<|im_end|>\n", "base_se_raw": [58.0, 62.0], "base_se_norm": [-0.4249878525733948, -0.38533180952072144], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.41507381200790405, -0.38533180952072144], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006859.pt"}
{"idx": 6861, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/V-zXT5bIBM0.mp4", "duration": 201.73486666666668, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the ship is being tossed around in a violent storm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 108.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the ship is being tossed around in a violent storm.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 108.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [108.0, 121.0], "base_se_norm": [0.07071232795715332, 0.1995943784713745], "gt_se_raw": [109.0, 115.0], "gt_se_norm": [0.08062624931335449, 0.14011037349700928], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006861.pt"}
{"idx": 6863, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bv2vT665bGI.mp4", "duration": 218.0, "prompt_text": "Please find the visual event described by the sentence 'What is the time interval for softening the lemon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'What is the time interval for softening the lemon?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.6788990497589111, -0.6238532066345215], "gt_se_raw": [35.0, 41.0], "gt_se_norm": [-0.6788990497589111, -0.6238532066345215], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006863.pt"}
{"idx": 6865, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Bv2vT665bGI.mp4", "duration": 218.0, "prompt_text": "Please find the visual event described by the sentence 'Show the process of connecting the copper clips and zinc nails with wires.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 85.0 - 118.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the process of connecting the copper clips and zinc nails with wires.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 85.0 - 118.0 seconds.<|im_end|>\n", "base_se_raw": [85.0, 118.0], "base_se_norm": [-0.22018349170684814, 0.08256876468658447], "gt_se_raw": [80.0, 117.0], "gt_se_norm": [-0.26605504751205444, 0.0733945369720459], "start_token_span": [68, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006865.pt"}
{"idx": 6867, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dxvFck5LBkU.mp4", "duration": 291.16, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put the baby in the high chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 40.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put the baby in the high chair?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 40.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 40.0], "base_se_norm": [-0.7595824003219604, -0.7252370119094849], "gt_se_raw": [35.0, 40.0], "gt_se_norm": [-0.7595824003219604, -0.7252370119094849], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006867.pt"}
{"idx": 6869, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dxvFck5LBkU.mp4", "duration": 291.16, "prompt_text": "Please find the visual event described by the sentence 'At what point is the woman placing the mixture onto the baking tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 124.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the woman placing the mixture onto the baking tray?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 124.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [124.0, 131.0], "base_se_norm": [-0.14823466539382935, -0.10015112161636353], "gt_se_raw": [124.0, 131.0], "gt_se_norm": [-0.14823466539382935, -0.10015112161636353], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006869.pt"}
{"idx": 6871, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/dxvFck5LBkU.mp4", "duration": 291.16, "prompt_text": "Please find the visual event described by the sentence 'Show me when the baby is eating the finished oat bite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 205.0 - 228.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the baby is eating the finished oat bite.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 205.0 - 228.0 seconds.<|im_end|>\n", "base_se_raw": [205.0, 228.0], "base_se_norm": [0.4081604480743408, 0.5661492347717285], "gt_se_raw": [212.0, 228.0], "gt_se_norm": [0.45624399185180664, 0.5661492347717285], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006871.pt"}
{"idx": 6873, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zNKWz8hRil4.mp4", "duration": 260.6770833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point do we see a marching band playing trumpets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 35.0 - 41.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do we see a marching band playing trumpets?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 35.0 - 41.0 seconds.<|im_end|>\n", "base_se_raw": [35.0, 41.0], "base_se_norm": [-0.7314685583114624, -0.6854345798492432], "gt_se_raw": [35.0, 37.0], "gt_se_norm": [-0.7314685583114624, -0.7161238789558411], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006873.pt"}
{"idx": 6875, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/zNKWz8hRil4.mp4", "duration": 260.6770833333333, "prompt_text": "Please find the visual event described by the sentence 'Show the moment when people are taking pictures with Scooby-Doo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 94.0 - 99.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the moment when people are taking pictures with Scooby-Doo.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 94.0 - 99.0 seconds.<|im_end|>\n", "base_se_raw": [94.0, 99.0], "base_se_norm": [-0.27880120277404785, -0.24043959379196167], "gt_se_raw": [95.0, 99.0], "gt_se_norm": [-0.27112889289855957, -0.24043959379196167], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006875.pt"}
{"idx": 6877, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kW1e1qTG_p4.mp4", "duration": 229.3291, "prompt_text": "Please find the visual event described by the sentence 'When are three people talking in front of a large metal nitrogen tank?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 24.0 - 29.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are three people talking in front of a large metal nitrogen tank?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 24.0 - 29.0 seconds.<|im_end|>\n", "base_se_raw": [24.0, 29.0], "base_se_norm": [-0.7906938195228577, -0.7470883131027222], "gt_se_raw": [25.0, 29.0], "gt_se_norm": [-0.7819727063179016, -0.7470883131027222], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006877.pt"}
{"idx": 6879, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kW1e1qTG_p4.mp4", "duration": 229.3291, "prompt_text": "Please find the visual event described by the sentence 'At what time do four people pose together for a photo outside a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do four people pose together for a photo outside a building?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 70.0], "base_se_norm": [-0.43312907218933105, -0.38952362537384033], "gt_se_raw": [66.0, 69.0], "gt_se_norm": [-0.424407958984375, -0.3982447385787964], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006879.pt"}
{"idx": 6881, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/kW1e1qTG_p4.mp4", "duration": 229.3291, "prompt_text": "Please find the visual event described by the sentence 'When does the camera show a close-up of a spinning yellow wheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 77.0 - 79.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the camera show a close-up of a spinning yellow wheel?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 77.0 - 79.0 seconds.<|im_end|>\n", "base_se_raw": [77.0, 79.0], "base_se_norm": [-0.3284759521484375, -0.31103378534317017], "gt_se_raw": [77.0, 78.0], "gt_se_norm": [-0.3284759521484375, -0.3197548985481262], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006881.pt"}
{"idx": 6883, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mVLrBJYGxk4.mp4", "duration": 303.55325, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the large robot enters the house for the daily check.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 62.0 - 69.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the large robot enters the house for the daily check.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 62.0 - 69.0 seconds.<|im_end|>\n", "base_se_raw": [62.0, 69.0], "base_se_norm": [-0.5915049314498901, -0.5453845262527466], "gt_se_raw": [63.0, 69.0], "gt_se_norm": [-0.584916353225708, -0.5453845262527466], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006883.pt"}
{"idx": 6885, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/mVLrBJYGxk4.mp4", "duration": 303.55325, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the old couple is being led away after being labeled 'UGLY'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the old couple is being led away after being labeled 'UGLY'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 182.0], "base_se_norm": [0.09371256828308105, 0.19913065433502197], "gt_se_raw": [167.0, 173.0], "gt_se_norm": [0.10030114650726318, 0.1398329734802246], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006885.pt"}
{"idx": 6887, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcIQiI_Yfg4.mp4", "duration": 250.25, "prompt_text": "Please find the visual event described by the sentence 'When can we see a map animation of the Hawaiian islands and volcanoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 72.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a map animation of the Hawaiian islands and volcanoes?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 72.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 72.0], "base_se_norm": [-0.4645354747772217, -0.42457544803619385], "gt_se_raw": [66.0, 72.0], "gt_se_norm": [-0.4725274443626404, -0.42457544803619385], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006887.pt"}
{"idx": 6889, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcIQiI_Yfg4.mp4", "duration": 250.25, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a man uses a hammer to take a sample from the cooled lava rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 110.0 - 114.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a man uses a hammer to take a sample from the cooled lava rock.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 110.0 - 114.0 seconds.<|im_end|>\n", "base_se_raw": [110.0, 114.0], "base_se_norm": [-0.12087911367416382, -0.08891111612319946], "gt_se_raw": [110.0, 111.0], "gt_se_norm": [-0.12087911367416382, -0.11288708448410034], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006889.pt"}
{"idx": 6891, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/OcIQiI_Yfg4.mp4", "duration": 250.25, "prompt_text": "Please find the visual event described by the sentence 'Show the shot where a group of photographers are set up on a cliff at dusk to capture the lava.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the shot where a group of photographers are set up on a cliff at dusk to capture the lava.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 131.0], "base_se_norm": [0.022976994514465332, 0.04695308208465576], "gt_se_raw": [129.0, 131.0], "gt_se_norm": [0.03096902370452881, 0.04695308208465576], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006891.pt"}
{"idx": 6893, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wBU9N35ZHIw.mp4", "duration": 308.12, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where an image of Camembert cheese is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where an image of Camembert cheese is displayed.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 34.0], "base_se_norm": [-0.8117616176605225, -0.7793067693710327], "gt_se_raw": [30.0, 34.0], "gt_se_norm": [-0.8052706718444824, -0.7793067693710327], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006893.pt"}
{"idx": 6895, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wBU9N35ZHIw.mp4", "duration": 308.12, "prompt_text": "Please find the visual event described by the sentence 'Find the moment when the man cuts the cake with green icing down the middle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 130.0 - 151.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment when the man cuts the cake with green icing down the middle.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 130.0 - 151.0 seconds.<|im_end|>\n", "base_se_raw": [130.0, 151.0], "base_se_norm": [-0.15617293119430542, -0.019862353801727295], "gt_se_raw": [130.0, 142.0], "gt_se_norm": [-0.15617293119430542, -0.07828116416931152], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006895.pt"}
{"idx": 6897, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihCciBFdbkw.mp4", "duration": 286.1859, "prompt_text": "Please find the visual event described by the sentence 'When is the Stan Lee cameo shown side-by-side with its live-action counterpart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the Stan Lee cameo shown side-by-side with its live-action counterpart?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 45.0], "base_se_norm": [-0.7064844965934753, -0.6855190992355347], "gt_se_raw": [43.0, 45.0], "gt_se_norm": [-0.6994960308074951, -0.6855190992355347], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006897.pt"}
{"idx": 6899, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihCciBFdbkw.mp4", "duration": 286.1859, "prompt_text": "Please find the visual event described by the sentence 'At what point does the video list the parody cast names for the characters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 220.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the video list the parody cast names for the characters?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 220.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 220.0], "base_se_norm": [0.42564666271209717, 0.5374621152877808], "gt_se_raw": [204.0, 220.0], "gt_se_norm": [0.42564666271209717, 0.5374621152877808], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006899.pt"}
{"idx": 6901, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ihCciBFdbkw.mp4", "duration": 286.1859, "prompt_text": "Please find the visual event described by the sentence 'Locate the clip where Hiro teaches Baymax a karate move by having him punch a wooden board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 109.0 - 112.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the clip where Hiro teaches Baymax a karate move by having him punch a wooden board.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 109.0 - 112.0 seconds.<|im_end|>\n", "base_se_raw": [109.0, 112.0], "base_se_norm": [-0.23825740814208984, -0.21729201078414917], "gt_se_raw": [109.0, 111.0], "gt_se_norm": [-0.23825740814208984, -0.2242804765701294], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006901.pt"}
{"idx": 6903, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgfvmHeOiqQ.mp4", "duration": 245.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment the heart-shaped creature is in the hand of the boy in the blue T-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 65.0 - 76.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment the heart-shaped creature is in the hand of the boy in the blue T-shirt.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 65.0 - 76.0 seconds.<|im_end|>\n", "base_se_raw": [65.0, 76.0], "base_se_norm": [-0.47109681367874146, -0.3815901279449463], "gt_se_raw": [66.0, 70.0], "gt_se_norm": [-0.46295982599258423, -0.4304119348526001], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006903.pt"}
{"idx": 6905, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/GgfvmHeOiqQ.mp4", "duration": 245.79166666666666, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the heart splits into two pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 160.0 - 163.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the heart splits into two pieces.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 160.0 - 163.0 seconds.<|im_end|>\n", "base_se_raw": [160.0, 163.0], "base_se_norm": [0.3019155263900757, 0.32632648944854736], "gt_se_raw": [161.0, 163.0], "gt_se_norm": [0.3100525140762329, 0.32632648944854736], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006905.pt"}
{"idx": 6907, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2JduAOOpOSI.mp4", "duration": 235.81891666666664, "prompt_text": "Please find the visual event described by the sentence 'When does the bald man wearing a tie walk towards his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 19.0 - 26.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the bald man wearing a tie walk towards his desk?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 19.0 - 26.0 seconds.<|im_end|>\n", "base_se_raw": [19.0, 26.0], "base_se_norm": [-0.8388594388961792, -0.7794918417930603], "gt_se_raw": [19.0, 26.0], "gt_se_norm": [-0.8388594388961792, -0.7794918417930603], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006907.pt"}
{"idx": 6909, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2JduAOOpOSI.mp4", "duration": 235.81891666666664, "prompt_text": "Please find the visual event described by the sentence 'Show me when the boss, wearing a white shirt and black tie, enters the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 91.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the boss, wearing a white shirt and black tie, enters the scene.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 91.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 91.0], "base_se_norm": [-0.2621457576751709, -0.22822141647338867], "gt_se_raw": [86.0, 90.0], "gt_se_norm": [-0.27062684297561646, -0.23670250177383423], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006909.pt"}
{"idx": 6911, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/2JduAOOpOSI.mp4", "duration": 235.81891666666664, "prompt_text": "Please find the visual event described by the sentence 'When can we see the close-up of the long-haired man playing a game on the retro phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 176.0 - 183.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see the close-up of the long-haired man playing a game on the retro phone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 176.0 - 183.0 seconds.<|im_end|>\n", "base_se_raw": [176.0, 183.0], "base_se_norm": [0.4926706552505493, 0.552038311958313], "gt_se_raw": [176.0, 182.0], "gt_se_norm": [0.4926706552505493, 0.5435571670532227], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006911.pt"}
{"idx": 6913, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K6M67cACkcQ.mp4", "duration": 321.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'Show the segment where the presenter clicks on the 'Restrictions' tab after selecting 'Edit Folder'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 63.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the segment where the presenter clicks on the 'Restrictions' tab after selecting 'Edit Folder'.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 63.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [63.0, 70.0], "base_se_norm": [-0.6078024506568909, -0.5642249584197998], "gt_se_raw": [59.0, 62.0], "gt_se_norm": [-0.6327039003372192, -0.6140277981758118], "start_token_span": [75, 79], "end_token_span": [81, 85], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006913.pt"}
{"idx": 6915, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/K6M67cACkcQ.mp4", "duration": 321.26666666666665, "prompt_text": "Please find the visual event described by the sentence 'Find the time range for navigating to the 'Quizzes' section via the 'Other Tools' menu.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 209.0 - 214.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the time range for navigating to the 'Quizzes' section via the 'Other Tools' menu.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 209.0 - 214.0 seconds.<|im_end|>\n", "base_se_raw": [209.0, 214.0], "base_se_norm": [0.3010997772216797, 0.33222663402557373], "gt_se_raw": [209.0, 212.0], "gt_se_norm": [0.3010997772216797, 0.31977593898773193], "start_token_span": [75, 80], "end_token_span": [82, 87], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006915.pt"}
{"idx": 6917, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/13S9n41lme8.mp4", "duration": 224.55766666666668, "prompt_text": "Please find the visual event described by the sentence 'When can we see a leopard shark swimming from right to left past the kelp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 16.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a leopard shark swimming from right to left past the kelp?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 16.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [16.0, 24.0], "base_se_norm": [-0.8574976325035095, -0.7862464189529419], "gt_se_raw": [18.0, 24.0], "gt_se_norm": [-0.8396848440170288, -0.7862464189529419], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006917.pt"}
{"idx": 6919, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/13S9n41lme8.mp4", "duration": 224.55766666666668, "prompt_text": "Please find the visual event described by the sentence 'Show the part where a diver is swimming amidst a school of fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 73.0 - 90.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where a diver is swimming amidst a school of fish.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 73.0 - 90.0 seconds.<|im_end|>\n", "base_se_raw": [73.0, 90.0], "base_se_norm": [-0.3498328924179077, -0.19842416048049927], "gt_se_raw": [74.0, 87.0], "gt_se_norm": [-0.34092652797698975, -0.22514331340789795], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006919.pt"}
{"idx": 6921, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/13S9n41lme8.mp4", "duration": 224.55766666666668, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when a horn shark is foraging for food in the rocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 222.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when a horn shark is foraging for food in the rocks.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 222.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 222.0], "base_se_norm": [0.620964527130127, 0.9772204160690308], "gt_se_raw": [193.0, 208.0], "gt_se_norm": [0.7189348936080933, 0.852530837059021], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006921.pt"}
{"idx": 6923, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LjufgOcZLZs.mp4", "duration": 273.24, "prompt_text": "Please find the visual event described by the sentence 'Show the interaction where a man in a suit takes the fifty-dollar bill and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 24.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the interaction where a man in a suit takes the fifty-dollar bill and walks away.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 24.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 24.0], "base_se_norm": [-0.9194846749305725, -0.8243302702903748], "gt_se_raw": [11.0, 21.0], "gt_se_norm": [-0.9194846749305725, -0.8462889790534973], "start_token_span": [72, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006923.pt"}
{"idx": 6925, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/LjufgOcZLZs.mp4", "duration": 273.24, "prompt_text": "Please find the visual event described by the sentence 'A man in a brown hat is eating ice cream. Show the time he appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 182.0 - 193.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a brown hat is eating ice cream. Show the time he appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 182.0 - 193.0 seconds.<|im_end|>\n", "base_se_raw": [182.0, 193.0], "base_se_norm": [0.3321622610092163, 0.412677526473999], "gt_se_raw": [183.0, 193.0], "gt_se_norm": [0.33948183059692383, 0.412677526473999], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006925.pt"}
{"idx": 6927, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4QdmRufojsU.mp4", "duration": 222.9562896229563, "prompt_text": "Please find the visual event described by the sentence 'When are people seen doing step aerobics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 13.0 - 19.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are people seen doing step aerobics?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 13.0 - 19.0 seconds.<|im_end|>\n", "base_se_raw": [13.0, 19.0], "base_se_norm": [-0.883385181427002, -0.8295630216598511], "gt_se_raw": [14.0, 19.0], "gt_se_norm": [-0.87441486120224, -0.8295630216598511], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006927.pt"}
{"idx": 6929, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4QdmRufojsU.mp4", "duration": 222.9562896229563, "prompt_text": "Please find the visual event described by the sentence 'At what point do two women use a resistance band together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 8.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do two women use a resistance band together?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 8.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [8.0, 13.0], "base_se_norm": [-0.9282370805740356, -0.883385181427002], "gt_se_raw": [68.0, 74.0], "gt_se_norm": [-0.3900149464607239, -0.3361927270889282], "start_token_span": [66, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006929.pt"}
{"idx": 6931, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/4QdmRufojsU.mp4", "duration": 222.9562896229563, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where individuals are lifting green dumbbells overhead.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 56.0 - 59.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where individuals are lifting green dumbbells overhead.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 56.0 - 59.0 seconds.<|im_end|>\n", "base_se_raw": [56.0, 59.0], "base_se_norm": [-0.4976593852043152, -0.47074824571609497], "gt_se_raw": [56.0, 59.0], "gt_se_norm": [-0.4976593852043152, -0.47074824571609497], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006931.pt"}
{"idx": 6933, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/98j7_2R5jdU.mp4", "duration": 320.0197, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where a person is climbing a tree using ropes, seen from a distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 83.0 - 93.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where a person is climbing a tree using ropes, seen from a distance.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 83.0 - 93.0 seconds.<|im_end|>\n", "base_se_raw": [83.0, 93.0], "base_se_norm": [-0.48128193616867065, -0.41878581047058105], "gt_se_raw": [84.0, 89.0], "gt_se_norm": [-0.47503232955932617, -0.44378429651260376], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006933.pt"}
{"idx": 6935, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/98j7_2R5jdU.mp4", "duration": 320.0197, "prompt_text": "Please find the visual event described by the sentence 'When is the man speaking inside a tent while it is raining outside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 7.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man speaking inside a tent while it is raining outside?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 7.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [7.0, 34.0], "base_se_norm": [-0.9562526941299438, -0.7875130772590637], "gt_se_raw": [12.0, 20.0], "gt_se_norm": [-0.9250046014785767, -0.875007688999176], "start_token_span": [68, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006935.pt"}
{"idx": 6937, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n7N6_lIORNM.mp4", "duration": 270.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When is the man sitting on the bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 0.0 - 3.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man sitting on the bench?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 0.0 - 3.0 seconds.<|im_end|>\n", "base_se_raw": [0.0, 3.0], "base_se_norm": [-1.0, -0.9778133630752563], "gt_se_raw": [0.0, 4.0], "gt_se_norm": [-1.0, -0.970417857170105], "start_token_span": [63, 66], "end_token_span": [68, 71], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006937.pt"}
{"idx": 6939, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n7N6_lIORNM.mp4", "duration": 270.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'At what point does a woman with red hair and glasses appear holding a bouquet of red flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 72.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a woman with red hair and glasses appear holding a bouquet of red flowers?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 72.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [72.0, 85.0], "base_se_norm": [-0.4675212502479553, -0.3713792562484741], "gt_se_raw": [72.0, 77.0], "gt_se_norm": [-0.4675212502479553, -0.43054354190826416], "start_token_span": [73, 77], "end_token_span": [79, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006939.pt"}
{"idx": 6941, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/n7N6_lIORNM.mp4", "duration": 270.43333333333334, "prompt_text": "Please find the visual event described by the sentence 'When can we see an aircraft flying across the sky above the statue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 255.0 - 259.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see an aircraft flying across the sky above the statue?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 255.0 - 259.0 seconds.<|im_end|>\n", "base_se_raw": [255.0, 259.0], "base_se_norm": [0.8858623504638672, 0.9154444932937622], "gt_se_raw": [255.0, 258.0], "gt_se_norm": [0.8858623504638672, 0.9080488681793213], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006941.pt"}
{"idx": 6943, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8j3NmTv9AWg.mp4", "duration": 252.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'Show me when the lemon falls asleep for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 70.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me when the lemon falls asleep for the first time.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 70.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 70.0], "base_se_norm": [-0.46076327562332153, -0.44490331411361694], "gt_se_raw": [69.0, 74.0], "gt_se_norm": [-0.45283329486846924, -0.4131835103034973], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006943.pt"}
{"idx": 6945, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8j3NmTv9AWg.mp4", "duration": 252.20833333333334, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when Grandpa Lemon is declared the winner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 171.0 - 174.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when Grandpa Lemon is declared the winner.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 171.0 - 174.0 seconds.<|im_end|>\n", "base_se_raw": [171.0, 174.0], "base_se_norm": [0.3560218811035156, 0.3798116445541382], "gt_se_raw": [172.0, 173.0], "gt_se_norm": [0.36395180225372314, 0.37188172340393066], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006945.pt"}
{"idx": 6947, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAyVk_sH42k.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the video show the 'BADLANDSCHUGS' logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 15.0 - 21.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the video show the 'BADLANDSCHUGS' logo?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 15.0 - 21.0 seconds.<|im_end|>\n", "base_se_raw": [15.0, 21.0], "base_se_norm": [-0.8506472110748291, -0.7909060716629028], "gt_se_raw": [17.0, 19.0], "gt_se_norm": [-0.8307334780693054, -0.8108198046684265], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006947.pt"}
{"idx": 6949, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAyVk_sH42k.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man pour the drink from the second plastic cup into the boot-shaped glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 95.0 - 104.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man pour the drink from the second plastic cup into the boot-shaped glass?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 95.0 - 104.0 seconds.<|im_end|>\n", "base_se_raw": [95.0, 104.0], "base_se_norm": [-0.05409890413284302, 0.03551280498504639], "gt_se_raw": [95.0, 104.0], "gt_se_norm": [-0.05409890413284302, 0.03551280498504639], "start_token_span": [72, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006949.pt"}
{"idx": 6951, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/KAyVk_sH42k.mp4", "duration": 200.86666666666667, "prompt_text": "Please find the visual event described by the sentence 'When does the man show the full, empty, and frosted boot-shaped glass to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 152.0 - 182.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man show the full, empty, and frosted boot-shaped glass to the camera?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 152.0 - 182.0 seconds.<|im_end|>\n", "base_se_raw": [152.0, 182.0], "base_se_norm": [0.5134416818618774, 0.8121473789215088], "gt_se_raw": [173.0, 182.0], "gt_se_norm": [0.7225356101989746, 0.8121473789215088], "start_token_span": [74, 79], "end_token_span": [81, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006951.pt"}
{"idx": 6953, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SbQc_JLUH7k.mp4", "duration": 200.375, "prompt_text": "Please find the visual event described by the sentence 'At what point is the man looking at himself in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 12.0 - 22.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the man looking at himself in the mirror?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 12.0 - 22.0 seconds.<|im_end|>\n", "base_se_raw": [12.0, 22.0], "base_se_norm": [-0.8802245855331421, -0.7804117202758789], "gt_se_raw": [11.0, 22.0], "gt_se_norm": [-0.8902058601379395, -0.7804117202758789], "start_token_span": [67, 71], "end_token_span": [73, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006953.pt"}
{"idx": 6955, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SbQc_JLUH7k.mp4", "duration": 200.375, "prompt_text": "Please find the visual event described by the sentence 'Identify the moment a hand reaches in and removes the man's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 82.0 - 84.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the moment a hand reaches in and removes the man's hair.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 82.0 - 84.0 seconds.<|im_end|>\n", "base_se_raw": [82.0, 84.0], "base_se_norm": [-0.18153464794158936, -0.16157203912734985], "gt_se_raw": [82.0, 84.0], "gt_se_norm": [-0.18153464794158936, -0.16157203912734985], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006955.pt"}
{"idx": 6957, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8VkD__hXeM8.mp4", "duration": 323.64, "prompt_text": "Please find the visual event described by the sentence 'When does the man in the suit take off his shoes before getting into bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man in the suit take off his shoes before getting into bed?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 36.0], "base_se_norm": [-0.8022494316101074, -0.7775306105613708], "gt_se_raw": [32.0, 36.0], "gt_se_norm": [-0.8022494316101074, -0.7775306105613708], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006957.pt"}
{"idx": 6959, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8VkD__hXeM8.mp4", "duration": 323.64, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man is flying horizontally through the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 115.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man is flying horizontally through the room.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 115.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [115.0, 130.0], "base_se_norm": [-0.28933387994766235, -0.196638286113739], "gt_se_raw": [117.0, 127.0], "gt_se_norm": [-0.2769744396209717, -0.21517741680145264], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006959.pt"}
{"idx": 6961, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/8VkD__hXeM8.mp4", "duration": 323.64, "prompt_text": "Please find the visual event described by the sentence 'When is the room being engulfed in a swirling brown liquid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 207.0 - 246.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the room being engulfed in a swirling brown liquid?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 207.0 - 246.0 seconds.<|im_end|>\n", "base_se_raw": [207.0, 246.0], "base_se_norm": [0.2791990041732788, 0.5202075242996216], "gt_se_raw": [207.0, 221.0], "gt_se_norm": [0.2791990041732788, 0.3657149076461792], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006961.pt"}
{"idx": 6963, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S5z5_ukDxII.mp4", "duration": 204.6044, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when a knot is tied at the end of the paper strip to form a pentagon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 42.0 - 78.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when a knot is tied at the end of the paper strip to form a pentagon.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 42.0 - 78.0 seconds.<|im_end|>\n", "base_se_raw": [42.0, 78.0], "base_se_norm": [-0.5894516706466675, -0.23755306005477905], "gt_se_raw": [42.0, 65.0], "gt_se_norm": [-0.5894516706466675, -0.36462754011154175], "start_token_span": [76, 80], "end_token_span": [82, 86], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006963.pt"}
{"idx": 6965, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/S5z5_ukDxII.mp4", "duration": 204.6044, "prompt_text": "Please find the visual event described by the sentence 'At what point is the flat paper pentagon being pinched to inflate it into a 3D star?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 145.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point is the flat paper pentagon being pinched to inflate it into a 3D star?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 145.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [145.0, 168.0], "base_se_norm": [0.41736936569213867, 0.6421934366226196], "gt_se_raw": [140.0, 164.0], "gt_se_norm": [0.36849451065063477, 0.6030936241149902], "start_token_span": [76, 81], "end_token_span": [83, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006965.pt"}
{"idx": 6967, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6azdnP1Dz0U.mp4", "duration": 212.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where a man with brown hair and a blue T-shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 11.0 - 27.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where a man with brown hair and a blue T-shirt is talking.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 11.0 - 27.0 seconds.<|im_end|>\n", "base_se_raw": [11.0, 27.0], "base_se_norm": [-0.896697461605072, -0.7464392185211182], "gt_se_raw": [12.0, 27.0], "gt_se_norm": [-0.8873063325881958, -0.7464392185211182], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006967.pt"}
{"idx": 6969, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6azdnP1Dz0U.mp4", "duration": 212.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'When is the toy train put into the elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 93.0 - 100.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the toy train put into the elevator?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 93.0 - 100.0 seconds.<|im_end|>\n", "base_se_raw": [93.0, 100.0], "base_se_norm": [-0.12662386894226074, -0.060885846614837646], "gt_se_raw": [92.0, 98.0], "gt_se_norm": [-0.13601499795913696, -0.07966816425323486], "start_token_span": [64, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006969.pt"}
{"idx": 6971, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/6azdnP1Dz0U.mp4", "duration": 212.96666666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where a man in a black cap with emojis on it appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 193.0 - 198.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where a man in a black cap with emojis on it appears.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 193.0 - 198.0 seconds.<|im_end|>\n", "base_se_raw": [193.0, 198.0], "base_se_norm": [0.8124902248382568, 0.8594459295272827], "gt_se_raw": [193.0, 199.0], "gt_se_norm": [0.8124902248382568, 0.8688371181488037], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006971.pt"}
{"idx": 6973, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/e9yPYbjA210.mp4", "duration": 277.277, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the man is signing consent forms at the reception.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 60.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the man is signing consent forms at the reception.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 60.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 60.0], "base_se_norm": [-0.6537758111953735, -0.5672197937965393], "gt_se_raw": [54.0, 57.0], "gt_se_norm": [-0.6104978322982788, -0.5888588428497314], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006973.pt"}
{"idx": 6975, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/e9yPYbjA210.mp4", "duration": 277.277, "prompt_text": "Please find the visual event described by the sentence 'Identify the time range where the man is running on the treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 167.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Identify the time range where the man is running on the treadmill.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 167.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [167.0, 188.0], "base_se_norm": [0.20457160472869873, 0.3560446500778198], "gt_se_raw": [168.0, 182.0], "gt_se_norm": [0.21178460121154785, 0.3127666711807251], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006975.pt"}
{"idx": 6977, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Slq2zeofV0w.mp4", "duration": 202.3021, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the text 'The Picnic' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 45.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the text 'The Picnic' on the screen?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 45.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 45.0], "base_se_norm": [-0.6638690233230591, -0.5551207661628723], "gt_se_raw": [34.0, 45.0], "gt_se_norm": [-0.6638690233230591, -0.5551207661628723], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006977.pt"}
{"idx": 6979, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Slq2zeofV0w.mp4", "duration": 202.3021, "prompt_text": "Please find the visual event described by the sentence 'When does the man step on the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 107.0 - 111.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the man step on the cake?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 107.0 - 111.0 seconds.<|im_end|>\n", "base_se_raw": [107.0, 111.0], "base_se_norm": [0.057823896408081055, 0.09736871719360352], "gt_se_raw": [108.0, 111.0], "gt_se_norm": [0.06771016120910645, 0.09736871719360352], "start_token_span": [63, 68], "end_token_span": [70, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006979.pt"}
{"idx": 6981, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Slq2zeofV0w.mp4", "duration": 202.3021, "prompt_text": "Please find the visual event described by the sentence 'Show me the scene where the couple in the movie embraces happily at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 166.0 - 172.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the scene where the couple in the movie embraces happily at the end.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 166.0 - 172.0 seconds.<|im_end|>\n", "base_se_raw": [166.0, 172.0], "base_se_norm": [0.6411100625991821, 0.7004272937774658], "gt_se_raw": [166.0, 171.0], "gt_se_norm": [0.6411100625991821, 0.6905410289764404], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006981.pt"}
{"idx": 6983, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/80zhPlAesEE.mp4", "duration": 312.28, "prompt_text": "Please find the visual event described by the sentence 'Show the shot where a cart filled with what appear to be bandaged bodies is being pulled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 9.0 - 13.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the shot where a cart filled with what appear to be bandaged bodies is being pulled.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 9.0 - 13.0 seconds.<|im_end|>\n", "base_se_raw": [9.0, 13.0], "base_se_norm": [-0.942359447479248, -0.9167413711547852], "gt_se_raw": [11.0, 13.0], "gt_se_norm": [-0.9295504093170166, -0.9167413711547852], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006983.pt"}
{"idx": 6985, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/80zhPlAesEE.mp4", "duration": 312.28, "prompt_text": "Please find the visual event described by the sentence 'Find the moment where a woman snatches a letter from another, starting a fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 139.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the moment where a woman snatches a letter from another, starting a fight.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 139.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [139.0, 142.0], "base_se_norm": [-0.10977327823638916, -0.09055972099304199], "gt_se_raw": [139.0, 146.0], "gt_se_norm": [-0.10977327823638916, -0.06494170427322388], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006985.pt"}
{"idx": 6987, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DS8yeXFeEPA.mp4", "duration": 323.65666666666664, "prompt_text": "Please find the visual event described by the sentence 'When are the students sitting tiredly at their desks before the exercise begins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 4.0 - 11.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the students sitting tiredly at their desks before the exercise begins?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 4.0 - 11.0 seconds.<|im_end|>\n", "base_se_raw": [4.0, 11.0], "base_se_norm": [-0.9752824306488037, -0.932026743888855], "gt_se_raw": [4.0, 11.0], "gt_se_norm": [-0.9752824306488037, -0.932026743888855], "start_token_span": [69, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006987.pt"}
{"idx": 6989, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DS8yeXFeEPA.mp4", "duration": 323.65666666666664, "prompt_text": "Please find the visual event described by the sentence 'At what point do they raise their hands and start jumping up and down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 67.0 - 88.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point do they raise their hands and start jumping up and down?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 67.0 - 88.0 seconds.<|im_end|>\n", "base_se_raw": [67.0, 88.0], "base_se_norm": [-0.585981011390686, -0.45621389150619507], "gt_se_raw": [67.0, 77.0], "gt_se_norm": [-0.585981011390686, -0.5241872072219849], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006989.pt"}
{"idx": 6991, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/DS8yeXFeEPA.mp4", "duration": 323.65666666666664, "prompt_text": "Please find the visual event described by the sentence 'Locate the time frame when everyone bends forward to stretch their backs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 285.0 - 292.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the time frame when everyone bends forward to stretch their backs.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 285.0 - 292.0 seconds.<|im_end|>\n", "base_se_raw": [285.0, 292.0], "base_se_norm": [0.7611254453659058, 0.8043811321258545], "gt_se_raw": [284.0, 292.0], "gt_se_norm": [0.7549459934234619, 0.8043811321258545], "start_token_span": [68, 73], "end_token_span": [75, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006991.pt"}
{"idx": 6993, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wWVppdfYOx8.mp4", "duration": 271.2, "prompt_text": "Please find the visual event described by the sentence 'When are the eyes and nose drawn on the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 247.0 - 262.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When are the eyes and nose drawn on the dog?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 247.0 - 262.0 seconds.<|im_end|>\n", "base_se_raw": [247.0, 262.0], "base_se_norm": [0.8215337991714478, 0.9321533441543579], "gt_se_raw": [245.0, 261.0], "gt_se_norm": [0.8067846298217773, 0.9247786998748779], "start_token_span": [65, 70], "end_token_span": [72, 77], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006993.pt"}
{"idx": 6995, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/wWVppdfYOx8.mp4", "duration": 271.2, "prompt_text": "Please find the visual event described by the sentence 'At what point are the sides of one of the flaps folded towards the center line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 100.0 - 130.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point are the sides of one of the flaps folded towards the center line?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 100.0 - 130.0 seconds.<|im_end|>\n", "base_se_raw": [100.0, 130.0], "base_se_norm": [-0.262536883354187, -0.041297972202301025], "gt_se_raw": [110.0, 130.0], "gt_se_norm": [-0.18879061937332153, -0.041297972202301025], "start_token_span": [72, 77], "end_token_span": [79, 84], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006995.pt"}
{"idx": 6997, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SmSVlY2gx88.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'When does the speaker first hold up the book 'Don't Call Me Special'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 21.0 - 33.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the speaker first hold up the book 'Don't Call Me Special'?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 21.0 - 33.0 seconds.<|im_end|>\n", "base_se_raw": [21.0, 33.0], "base_se_norm": [-0.8114245533943176, -0.703667163848877], "gt_se_raw": [22.0, 33.0], "gt_se_norm": [-0.8024447560310364, -0.703667163848877], "start_token_span": [71, 75], "end_token_span": [77, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006997.pt"}
{"idx": 6999, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SmSVlY2gx88.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'Find the animated clip where a disabled boy is excluded from a soccer game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 70.0 - 83.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the animated clip where a disabled boy is excluded from a soccer game.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 70.0 - 83.0 seconds.<|im_end|>\n", "base_se_raw": [70.0, 83.0], "base_se_norm": [-0.3714151382446289, -0.2546778917312622], "gt_se_raw": [70.0, 85.0], "gt_se_norm": [-0.3714151382446289, -0.23671835660934448], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00006999.pt"}
{"idx": 7001, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/SmSVlY2gx88.mp4", "duration": 222.7225, "prompt_text": "Please find the visual event described by the sentence 'Show the summary slide that lists the key messages of the presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 208.0 - 218.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the summary slide that lists the key messages of the presentation.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 208.0 - 218.0 seconds.<|im_end|>\n", "base_se_raw": [208.0, 218.0], "base_se_norm": [0.8677951097488403, 0.9575929641723633], "gt_se_raw": [208.0, 218.0], "gt_se_norm": [0.8677951097488403, 0.9575929641723633], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007001.pt"}
{"idx": 7003, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JfvEPc07UVY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'Locate the moment when students are receiving their diplomas at a graduation ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the moment when students are receiving their diplomas at a graduation ceremony.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.45358508825302124, -0.4121900200843811], "gt_se_raw": [66.0, 71.0], "gt_se_norm": [-0.45358508825302124, -0.4121900200843811], "start_token_span": [70, 74], "end_token_span": [76, 80], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007003.pt"}
{"idx": 7005, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JfvEPc07UVY.mp4", "duration": 241.57466666666667, "prompt_text": "Please find the visual event described by the sentence 'When can we see a group of women having a cooking class?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 142.0 - 145.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we see a group of women having a cooking class?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 142.0 - 145.0 seconds.<|im_end|>\n", "base_se_raw": [142.0, 145.0], "base_se_norm": [0.1756199598312378, 0.20045697689056396], "gt_se_raw": [142.0, 145.0], "gt_se_norm": [0.1756199598312378, 0.20045697689056396], "start_token_span": [67, 72], "end_token_span": [74, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007005.pt"}
{"idx": 7007, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1Ez6dw3ywcc.mp4", "duration": 207.70749999999998, "prompt_text": "Please find the visual event described by the sentence 'When can we observe the man in the light-colored shirt making a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 32.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When can we observe the man in the light-colored shirt making a phone call?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 32.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 32.0], "base_se_norm": [-0.9422264695167542, -0.6918743848800659], "gt_se_raw": [26.0, 40.0], "gt_se_norm": [-0.7496479749679565, -0.6148430109024048], "start_token_span": [70, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007007.pt"}
{"idx": 7009, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1Ez6dw3ywcc.mp4", "duration": 207.70749999999998, "prompt_text": "Please find the visual event described by the sentence 'At what point does the man violently grab the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 87.0 - 95.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does the man violently grab the woman?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 87.0 - 95.0 seconds.<|im_end|>\n", "base_se_raw": [87.0, 95.0], "base_se_norm": [-0.1622835397720337, -0.08525210618972778], "gt_se_raw": [88.0, 95.0], "gt_se_norm": [-0.15265458822250366, -0.08525210618972778], "start_token_span": [65, 69], "end_token_span": [71, 75], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007009.pt"}
{"idx": 7011, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/1Ez6dw3ywcc.mp4", "duration": 207.70749999999998, "prompt_text": "Please find the visual event described by the sentence 'Find the scene where the woman holds her hands behind her back, displaying a ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 165.0 - 168.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Find the scene where the woman holds her hands behind her back, displaying a ring.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 165.0 - 168.0 seconds.<|im_end|>\n", "base_se_raw": [165.0, 168.0], "base_se_norm": [0.5887726545333862, 0.6176594495773315], "gt_se_raw": [166.0, 168.0], "gt_se_norm": [0.5984015464782715, 0.6176594495773315], "start_token_span": [71, 76], "end_token_span": [78, 83], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007011.pt"}
{"idx": 7013, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vqDbMEdLiCs.mp4", "duration": 238.96, "prompt_text": "Please find the visual event described by the sentence 'Show the part where the man in the green hoodie smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 32.0 - 34.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the part where the man in the green hoodie smiles.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 32.0 - 34.0 seconds.<|im_end|>\n", "base_se_raw": [32.0, 34.0], "base_se_norm": [-0.7321727275848389, -0.7154335975646973], "gt_se_raw": [32.0, 34.0], "gt_se_norm": [-0.7321727275848389, -0.7154335975646973], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007013.pt"}
{"idx": 7015, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vqDbMEdLiCs.mp4", "duration": 238.96, "prompt_text": "Please find the visual event described by the sentence 'Show the close-up of the ice cube melting rapidly on the aluminum block.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 135.0 - 142.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the close-up of the ice cube melting rapidly on the aluminum block.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 135.0 - 142.0 seconds.<|im_end|>\n", "base_se_raw": [135.0, 142.0], "base_se_norm": [0.1298961639404297, 0.1884833574295044], "gt_se_raw": [141.0, 143.0], "gt_se_norm": [0.1801137924194336, 0.19685304164886475], "start_token_span": [69, 74], "end_token_span": [76, 81], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007015.pt"}
{"idx": 7017, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ikGVWEvUzNM.mp4", "duration": 318.2083333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the woman put on a mask in the bathroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 29.0 - 38.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the woman put on a mask in the bathroom?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 29.0 - 38.0 seconds.<|im_end|>\n", "base_se_raw": [29.0, 38.0], "base_se_norm": [-0.8177294731140137, -0.7611627578735352], "gt_se_raw": [29.0, 32.0], "gt_se_norm": [-0.8177294731140137, -0.7988739013671875], "start_token_span": [66, 70], "end_token_span": [72, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007017.pt"}
{"idx": 7019, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ikGVWEvUzNM.mp4", "duration": 318.2083333333333, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the teacher is writing 'Plato's Allegory of the cave' on the whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 76.0 - 85.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the teacher is writing 'Plato's Allegory of the cave' on the whiteboard.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 76.0 - 85.0 seconds.<|im_end|>\n", "base_se_raw": [76.0, 85.0], "base_se_norm": [-0.5223255157470703, -0.4657588005065918], "gt_se_raw": [78.0, 83.0], "gt_se_norm": [-0.5097551345825195, -0.4783291816711426], "start_token_span": [78, 82], "end_token_span": [84, 88], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007019.pt"}
{"idx": 7021, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/ikGVWEvUzNM.mp4", "duration": 318.2083333333333, "prompt_text": "Please find the visual event described by the sentence 'When is a person in a white mask and scarf drawing in a notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 66.0 - 71.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a person in a white mask and scarf drawing in a notebook?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 66.0 - 71.0 seconds.<|im_end|>\n", "base_se_raw": [66.0, 71.0], "base_se_norm": [-0.5851774215698242, -0.5537514686584473], "gt_se_raw": [67.0, 71.0], "gt_se_norm": [-0.5788922309875488, -0.5537514686584473], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007021.pt"}
{"idx": 7023, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTb-DIU9vj8.mp4", "duration": 267.6840833333333, "prompt_text": "Please find the visual event described by the sentence 'At what point does a girl whisper something to her friend on the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 34.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what point does a girl whisper something to her friend on the playground?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 34.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [34.0, 36.0], "base_se_norm": [-0.7459691762924194, -0.7310262322425842], "gt_se_raw": [34.0, 37.0], "gt_se_norm": [-0.7459691762924194, -0.7235547304153442], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007023.pt"}
{"idx": 7025, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/lTb-DIU9vj8.mp4", "duration": 267.6840833333333, "prompt_text": "Please find the visual event described by the sentence 'When is the boy sitting alone in the office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 68.0 - 74.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the boy sitting alone in the office?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 68.0 - 74.0 seconds.<|im_end|>\n", "base_se_raw": [68.0, 74.0], "base_se_norm": [-0.49193841218948364, -0.4471094608306885], "gt_se_raw": [67.0, 72.0], "gt_se_norm": [-0.49940991401672363, -0.46205246448516846], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007025.pt"}
{"idx": 7027, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JuKCOthud68.mp4", "duration": 217.3838333333333, "prompt_text": "Please find the visual event described by the sentence 'When is the man playing the drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 185.0 - 188.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is the man playing the drums?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 185.0 - 188.0 seconds.<|im_end|>\n", "base_se_raw": [185.0, 188.0], "base_se_norm": [0.7020584344863892, 0.7296594381332397], "gt_se_raw": [185.0, 188.0], "gt_se_norm": [0.7020584344863892, 0.7296594381332397], "start_token_span": [62, 67], "end_token_span": [69, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007027.pt"}
{"idx": 7029, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JuKCOthud68.mp4", "duration": 217.3838333333333, "prompt_text": "Please find the visual event described by the sentence 'At what time do we see the children's choir singing on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 22.0 - 25.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'At what time do we see the children's choir singing on stage?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 22.0 - 25.0 seconds.<|im_end|>\n", "base_se_raw": [22.0, 25.0], "base_se_norm": [-0.7975930571556091, -0.7699921131134033], "gt_se_raw": [23.0, 25.0], "gt_se_norm": [-0.7883927226066589, -0.7699921131134033], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007029.pt"}
{"idx": 7031, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/JuKCOthud68.mp4", "duration": 217.3838333333333, "prompt_text": "Please find the visual event described by the sentence 'When does the blonde woman speak into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 125.0 - 128.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When does the blonde woman speak into the microphone?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 125.0 - 128.0 seconds.<|im_end|>\n", "base_se_raw": [125.0, 128.0], "base_se_norm": [0.15003955364227295, 0.17764043807983398], "gt_se_raw": [125.0, 127.0], "gt_se_norm": [0.15003955364227295, 0.1684401035308838], "start_token_span": [64, 69], "end_token_span": [71, 76], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007031.pt"}
{"idx": 7033, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vHjv17F8iiE.mp4", "duration": 201.9, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where 'VICTORY!' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 128.0 - 131.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where 'VICTORY!' is displayed on the screen.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 128.0 - 131.0 seconds.<|im_end|>\n", "base_se_raw": [128.0, 131.0], "base_se_norm": [0.2679544687271118, 0.2976721525192261], "gt_se_raw": [129.0, 131.0], "gt_se_norm": [0.2778604030609131, 0.2976721525192261], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007033.pt"}
{"idx": 7035, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/vHjv17F8iiE.mp4", "duration": 201.9, "prompt_text": "Please find the visual event described by the sentence 'When is a player seen falling into the void?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 36.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When is a player seen falling into the void?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 36.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 36.0], "base_se_norm": [-0.7325408458709717, -0.6433877944946289], "gt_se_raw": [26.0, 28.0], "gt_se_norm": [-0.742446780204773, -0.7226349711418152], "start_token_span": [64, 68], "end_token_span": [70, 74], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007035.pt"}
{"idx": 7037, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFWmYhkMB5Y.mp4", "duration": 327.16016666666667, "prompt_text": "Please find the visual event described by the sentence 'When do two men look at a photo of a man dressed as a sailor on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 6.0 - 28.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do two men look at a photo of a man dressed as a sailor on a tablet?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 6.0 - 28.0 seconds.<|im_end|>\n", "base_se_raw": [6.0, 28.0], "base_se_norm": [-0.9633207321166992, -0.8288300037384033], "gt_se_raw": [20.0, 22.0], "gt_se_norm": [-0.8777357339859009, -0.8655092716217041], "start_token_span": [73, 76], "end_token_span": [78, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007037.pt"}
{"idx": 7039, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFWmYhkMB5Y.mp4", "duration": 327.16016666666667, "prompt_text": "Please find the visual event described by the sentence 'A man in a tree costume captures another man in a suit and white mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 204.0 - 209.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'A man in a tree costume captures another man in a suit and white mask.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 204.0 - 209.0 seconds.<|im_end|>\n", "base_se_raw": [204.0, 209.0], "base_se_norm": [0.24709558486938477, 0.27766168117523193], "gt_se_raw": [206.0, 208.0], "gt_se_norm": [0.25932204723358154, 0.2715485095977783], "start_token_span": [70, 75], "end_token_span": [77, 82], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007039.pt"}
{"idx": 7041, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/pFWmYhkMB5Y.mp4", "duration": 327.16016666666667, "prompt_text": "Please find the visual event described by the sentence 'When do the two men scream in fear at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 270.0 - 273.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'When do the two men scream in fear at the end?', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 270.0 - 273.0 seconds.<|im_end|>\n", "base_se_raw": [270.0, 273.0], "base_se_norm": [0.6505677700042725, 0.6689074039459229], "gt_se_raw": [271.0, 273.0], "gt_se_norm": [0.6566809415817261, 0.6689074039459229], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007041.pt"}
{"idx": 7043, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g0FMrjsYAzc.mp4", "duration": 318.61830000000003, "prompt_text": "Please find the visual event described by the sentence 'Show the first-person perspective of someone getting out of a red helicopter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 27.0 - 48.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show the first-person perspective of someone getting out of a red helicopter.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 27.0 - 48.0 seconds.<|im_end|>\n", "base_se_raw": [27.0, 48.0], "base_se_norm": [-0.8305181860923767, -0.6986989974975586], "gt_se_raw": [28.0, 39.0], "gt_se_norm": [-0.8242411017417908, -0.7551929950714111], "start_token_span": [68, 72], "end_token_span": [74, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007043.pt"}
{"idx": 7045, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/g0FMrjsYAzc.mp4", "duration": 318.61830000000003, "prompt_text": "Please find the visual event described by the sentence 'Show me the part where the blue helicopter is taking off.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 213.0 - 218.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Show me the part where the blue helicopter is taking off.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 213.0 - 218.0 seconds.<|im_end|>\n", "base_se_raw": [213.0, 218.0], "base_se_norm": [0.3370230197906494, 0.36840856075286865], "gt_se_raw": [214.0, 219.0], "gt_se_norm": [0.3433002233505249, 0.3746856451034546], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007045.pt"}
{"idx": 7047, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mprr5Q5Z7H4.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'Pinpoint the time when the pasta is being added to the boiling water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 48.0 - 65.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Pinpoint the time when the pasta is being added to the boiling water.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 48.0 - 65.0 seconds.<|im_end|>\n", "base_se_raw": [48.0, 65.0], "base_se_norm": [-0.7087732553482056, -0.6056303977966309], "gt_se_raw": [48.0, 65.0], "gt_se_norm": [-0.7087732553482056, -0.6056303977966309], "start_token_span": [69, 73], "end_token_span": [75, 79], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007047.pt"}
{"idx": 7049, "status": "ok", "video_file": "/mnt/localssd/project2/vlmversecast/data/videos/queryd/Mprr5Q5Z7H4.mp4", "duration": 329.64, "prompt_text": "Please find the visual event described by the sentence 'Locate the scene where butter is melting in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.", "rollout_text": "The event happens in 111.0 - 121.0 seconds.", "rendered_text": "<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>Please find the visual event described by the sentence 'Locate the scene where butter is melting in the pan.', determining its starting and ending times. The format should be: 'The event happens in <start time> - <end time> seconds'.<|im_end|>\n<|im_start|>assistant\nThe event happens in 111.0 - 121.0 seconds.<|im_end|>\n", "base_se_raw": [111.0, 121.0], "base_se_norm": [-0.3265380859375, -0.2658658027648926], "gt_se_raw": [111.0, 118.0], "gt_se_norm": [-0.3265380859375, -0.2840675115585327], "start_token_span": [66, 71], "end_token_span": [73, 78], "drop_reason": null, "tensor_path": "/mnt/localssd/project2/vlmversecast/cache/queryd/records/sample_00007049.pt"}